Meteorologic and Air Pollution Data
A description of the 24-hour averages of meteorologic and air pollution variables for each monitoring site can be found in eTable 1 (http://links.lww.com/EDE/A490). Medians of these site-specific descriptive measures are shown in Table 2. Mean temperature in summer was 16.5°C and in winter 5.4°C.
Air Temperature and Blood Pressure
A 10°C decrease in temperature was associated with an immediate (lag 0) increase in systolic BP by 0.5% (95% confidence interval [CI = 0.1% to 1.0%]) and a delayed (lag 6) increase in systolic BP by 0.4% (0.0% to 0.9%) (Fig. 1A). These percent changes correspond to an immediate increase in systolic BP of 0.6 mm Hg (0.1 to 1.1 mm Hg) and a delayed increase of 0.5 mm Hg (0.0 to 1.0 mm Hg) in association with a 10°C decrease in temperature. When we additionally adjusted for NO2, the sample size decreased by about 10% but temperature effects were stronger for lags 1, 2, and 5 and for the 7-day average (Fig. 1A). Temperature effects on systolic BP were not affected by adjustment for PM10 (data not shown). For all time lags, temperature effects were much stronger in April–September than in October–March (Fig. 1B). We observed no main effects of temperature on diastolic BP, but interaction analyses revealed a statistically significant increase in diastolic BP in association with temperature decrements in summer (data not shown). Temperature effects were not modified by center, smoking, or trimester of pregnancy.
Air Pollution and Blood Pressure
The medians of the station-specific interquartile ranges for 24-hour and 7-day averages were 14.4 and 11.4 μg/m3, respectively, for NO2 and 11.3 and 7.7 μg/m3 for PM10, respectively. Figure 2A shows the associations between NO2 and systolic BP, adjusted for the variables listed in eTable 2 (http://links.lww.com/EDE/A490). An IQR increase in NO2 in the 24 hours preceding the visit was associated with a change in systolic BP of −0.3% (−0.5% to −0.1%) or −0.4 mm Hg (−0.6 to −0.1 mm Hg). Similar decreases were also observed with lags of 1, 5, and 6 days. Increases in the 7-day average for both NO2 (−0.4% [−0.7% to −0.2%] or −0.5 mm Hg [0.8% to −0.2% mm Hg]) and PM10 (−0.3% [−0.5% to 0.0%] or −0.3 mm Hg [−0.6% to 0.0%]) were associated with the strongest reductions in SBP. Effect estimates for NO2 on systolic BP seemed to be stronger than PM10 effects in analyses considering pregnancy as a whole. We detected no clear associations between air pollutants and diastolic BP (Fig. 2B).
Effect Modification of Air Pollution Effects
Air pollution effect estimates were similar for the 2 cities (eFigure 2, http://links.lww.com/EDE/A490). There was a consistent pattern of effect modification by season, with a tendency to weaker negative effects of air pollutants in the April–September, compared with the October–March periods (Fig. 3A–3B). This effect modification reached statistical significance only for PM10 concentrations with a 2-day lag (−0.3% [−0.7% to −0.0%] in October–March versus 0.1% [−0.2% to 0.5%] in April–September, P value of interaction = 0.04). NO2 effects on systolic BP tended to be more pronounced in nonsmoking compared with smoking women (lowest P value of interaction = 0.12; eFigure 3, http://links.lww.com/EDE/A490).
We detected a strong modification of PM10, but not of NO2 effects, on systolic BP by trimester of pregnancy (Fig. 3C–D). The strongest modification was observed with the 4-day lag, which showed a 1.0% (0.5% to 1.5%) change in systolic BP associated with PM10 during the first trimester and changes of −0.3% (−0.6% to 0.0%) and −0.2% (−0.6% to 0.2%) during the second and third trimesters, respectively (P value of interaction <0.001).
Increases in PM10 (lag 0) were associated with a positive change in diastolic BP during the first trimester (1.1% [0.3% to 1.8%]), little change during the second trimester (0.2% [−0.3% to 0.8%]), and a negative change during the third trimester (−0.5% [−0.9% to 0.0%]).
Air pollution effects were similar when we included meteorologic variables with the same lag as the analyzed air pollution lag, after the exclusion of visits during the weekend or of the 46 women who developed gestational hypertension (data not shown). NO2 effects (lags 0, 1, and 2) slightly strengthened when we reduced the study population to women living ≤10 km from the closest monitoring station (NO2 effects with a lag of 0, 1, and 2 days were −0.4% [−0.6% to −1.5%], −0.3% [−0.6% to −0.1%], and −0.3% [−0.5% to −0.0%], respectively) or within 5 km (−0.4% [−0.7% to −0.2%], −0.4% [−0.7% to −0.1%], and −0.4% [−0.6% to −0.1%]) (eFigure 4A, http:links.lww.com/EDE/A490). When we included only the 220 women (with 1439 systolic BP measurements) living within a radius of 1 km to the closest NO2 station, no associations were observed and CIs were considerably widened. PM10 effect estimates did not change when we included only women living closer to the next monitoring station, but CIs widened (not shown). The air pollution effects on systolic BP were not altered when we used a spatial instead of a compound symmetry covariance structure. In the case of diastolic BP, convergence problems occurred, and a comparison between the models with different covariance structures was not possible. Using averaged air pollution measurements of several monitoring stations resulted in a smaller sample size but similar air pollution effects (eFigure 4B, http://links.lww.com/EDE/A490). Dispersion model estimates were available for 1155 women (eFigure 5, http://links.lww.com/EDE/A490). Seven-day averages of NO2 estimated with the dispersion model exhibited a similar (somewhat weaker) association with BP than when estimated with the closest station approach, with wider CIs. When the entire duration of pregnancy was considered, the effect of PM10 on systolic BP was stronger with the dispersion approach compared with the closest-station approach (eFigure 6, http://links.lww.com/EDE/A490). We found no interaction between trimester and PM10 when using the dispersion model (P value of interaction = 0.37, compared with P < 0.001 using the closest-station approach).
In our cohort of healthy pregnant women with no previous history of hypertension, we observed short-term reductions in systolic BP in association with increasing NO2 levels. This association was detected in all 3 trimesters of pregnancy. PM10 concentrations were positively associated with systolic and diastolic BP in the first trimester of pregnancy and tended to be negatively associated with both measures of BP later in pregnancy. This was not found with the alternative (dispersion) exposure model. We detected a short-term increase in systolic BP associated with decrements in temperature. These temperature effects persisted after adjustment for NO2, indicating at least partially independent associations of temperature and air pollution with systolic BP. The lack of associations between either temperature or air pollution and diastolic BP might be because diastolic BP is less accurately assessed than systolic BP.
Air Temperature and Blood Pressure
Elderly participants exhibit a higher BP in winter than in summer.13,21 We observed a slightly higher systolic BP in winter among pregnant women; air temperature effects on BP were more pronounced in summer. Among adults aged 35–64 years, Barnett et al14 reported an increase in average systolic BP of 0.19 mm Hg in association with a 1°C decrease in temperature. We also observed an immediate but weaker increase in systolic BP of 0.06 mm Hg with a 1°C decrease in temperature. Halonen et al22 reported increases in systolic BP of 0.6%–1.3% associated with decreases in apparent temperature among elderly men living in Boston. Furthermore, they detected elevated diastolic BP levels associated with decreases in air temperature and apparent temperature. However, in these studies, effects of apparent temperature might also partly reflect the association between relative humidity (or air pollution) and BP. Temperature effects in our study were weaker, and we did not find an association with diastolic BP.
Seasonal variation in BP might be explained by changes in blood viscosity favoring a decreased systemic pressure in summer compared with colder seasons.23 Furthermore, exposure to cold temperatures may activate the sympathetic nervous system and increase secretion of catecholamine. This possibly results in an increased heart rate and peripheral vascular resistance, and thus to increased BP with lower temperature.13,23
Air Pollution and Blood Pressure
The association between air pollution and BP has been characterized mainly in elderly participants5,24,25 and in people with underlying cardiovascular diseases.3,4,7 Findings are not consistent, perhaps due to misclassification of BP, differing confounder adjustment, or random associations because of multiple testing. Exposure misclassification and diversity in chemical composition of PM might also contribute to the inconsistent findings. Accordingly, Brook et al26 found no association of BP with PM2.5 estimated by ambient monitors, but did with personally measured PM2.5 levels. Only one study, based on Generation R cohort,8 has investigated the influence of air pollution on BP in pregnant women. BP was estimated once per trimester and exposure was averaged over each trimester of pregnancy. The Generation R study contrasts with ours by its focus on longer-term effects of air pollution, and on between-subject rather than within-subject variations in exposure. Van den Hooven and colleagues8 observed a positive association between trimester averages of PM10 and systolic BP only in the second and third trimesters of pregnancy; we observed a positive association with 1-day to 1-week PM10 averages only in the first trimester. Since the PM10 effect based on our dispersion model was not modified by trimester of pregnancy, the increase in systolic BP associated with first trimester PM10 levels from the closest station in our study should be considered with caution. Disregarding issues related to exposure misclassification, the gradual and profound changes in cardiovascular function during pregnancy27 make it plausible that environmental stressors could have different effects during different trimesters of pregnancy. Although we detected negative short-term NO2 effects on systolic BP, van den Hooven et al8 found an elevated systolic BP in association with trimester-specific NO2 increases. Differences in time scale and design make comparisons between these 2 studies difficult; taken together (and assuming that these findings cannot be explained by biases), these studies suggest that air pollutants could have different effects on the shorter- and longer-terms. Differences such as those between estimated effects of NO2 and PM10 are common in air pollution epidemiology. Such differences might be explained by the fact that these pollutants capture different dimensions of atmospheric pollutants, with PM10 coming in primarily from long-range pollution while NO2 levels come from predominantly local sources.
A positive association between PM and BP could be mediated by an activation of the sympathetic nervous system due to a stimulation of nerve endings in the human airways by inhaled particles.2 Furthermore, PM might trigger a systemic inflammation and oxidative reactions promoting vascular dysfunction.2 An inhalation chamber study with 27 participants found an increase in the endothelium-dependent vasoconstrictor endothelin-1 induced by diesel exhaust particles.28 Ultrafine particles, which are usually highly correlated with NO2, can pass alveolar walls and might directly influence endothelial cell structure and endothelial function, possibly leading to vasoconstriction and increased BP.2,29
Other authors have raised alternative hypotheses to explain a negative association between air pollution levels and BP. Cheng et al,30 who found a decreased BP in hypertensive rats exposed to concentrated particles, suggested that particles may cause airway irritation leading to increases of parasympathetic tone of the heart and peripheral vascular system. This hypothesis is supported by Zareba et al,31 who exposed 12 participants to ultrafine particles and observed such changes in electrocardiogram parameters as QT-shortening and ST-elevation, indicating an increase in parasympathetic tone.
Whether one of these mechanisms could apply to pregnant women remains to be investigated. Indeed, pregnant women are a very specific population. Pregnancy-related changes include hormonal changes, increases in blood volume and heart rate over the course of pregnancy, and decreasing BP in the second trimester of pregnancy.27 For these reasons, pregnant women may have different susceptibility to air pollution than nonpregnant and elderly subjects, and our results cannot be generalized to the whole population. Additionally, these considerable cardiac and hormonal changes throughout pregnancy could modify air pollution effects throughout pregnancy, as supported by the possible effect modification of PM10 by trimester of pregnancy in our population.
Changes in BP might act as an intermediate step between exposure to air pollution and adverse pregnancy outcomes, such as preterm birth and reduced fetal growth.11 Warland and McCutcheon32 reviewed the association between hypotension and poor pregnancy outcomes and concluded that low BP might result in preterm birth, perinatal mortality, and low birth weight. However, Zhang and Klebanoff33 suggested that the association between low BP during pregnancy and poor perinatal outcomes is due to confounding by other risk factors. Consequently, it is unclear whether the decreases in systolic BP reported in our study in association with air pollution could affect pregnancy outcome. High BP during pregnancy is a risk factor for preterm birth and low birth weight,34,35 and an increase in systolic BP in the first trimester associated with PM10 might have consequences for pregnancy outcomes.
Strengths and Limitations
The main strength of our study is the large number of women who participated in repeated BP measurements. This enabled us to analyze intraindividual variations in BP, avoiding bias due to potential confounders constant over time. A further strength is the availability of clinical characteristics, allowing us to efficiently control for these variables and perform subgroup analyses. BP is a highly variable parameter which is affected by such characteristics as age, medication, smoking, and weather. However, we could control our models for most of these factors. Moreover, we performed several sensitivity analyses that did not change our findings. A limitation is the lack of information on the exact times of BP measurements. BP has its own diurnal rhythms. This is unlikely to have induced confounding but may have caused exposure misclassification. The extent of misclassification is probably limited for the 7-day exposure averages, because these averages are unlikely to differ strongly if they end at 8 am or, say, 3 pm on the visit day. Only ambient exposure was measured, although people usually spend a lot of time indoors. NO2 effects tended to vary according to the buffer size considered around monitoring stations. Because population size varied with buffer size, the relative contribution of exposure misclassification and selection effects to these variations in estimated NO2 effects cannot easily be determined.
In conclusion, we observed a reduced systolic BP in pregnant women in association with elevated levels of NO2 and third trimester PM10 levels, and an increased systolic BP with higher PM10 levels during the first trimester of pregnancy. The consequences of such BP effects on pregnancy-related outcomes require further investigations. Additionally, our study observed decreases in systolic BP with increasing temperatures.
We thank the midwife research assistants (L. Douhaud, S. Bedel, B. Lortholary, S. Gabriel, M. Rogeon, and M. Malinbaum) for data collection; P. Lavoine for checking, coding, and data entry; J. Cyrys and S. von Klot for advice in exposure assessment; and S. Breitner for support in statistical analyses. We also thank Agnès Hulin, Fabrice Caïni (Atmo Poitou-Charentes), and Julien Galineau (Airlor) for the implementation of the dispersion model; and Lise Giorgis-Allemand (INSERM) for her statistical help.
1.Pope CAI, Dockery DW. Health effects of fine particulate air pollution: lines that connect. J Air Waste Manag Assoc
2.Brook RD, Rajagopalan S, Pope CA III, et al. Particulate matter air pollution and cardiovascular disease: an update to the scientific statement from the American Heart Association. Circulation
3.Ibald-Mulli A, Timonen KL, Peters A, et al. Effects of particulate air pollution on blood pressure and heart rate in subjects with cardiovascular disease: a multicenter approach. Environ Health Perspect
4.Delfino RJ, Tjoa T, Gillen DL, et al. Traffic-related air pollution and blood pressure in elderly subjects with coronary artery disease. Epidemiology
5.Harrabi I, Rondeau V, Dartigues JF, Tessier JF, Filleul L. Effects of particulate air pollution on systolic blood pressure: a population-based approach. Environ Res
6.Auchincloss AH, Diez Roux AV, Dvonch JT, et al. Associations between recent exposure to ambient fine particulate matter and blood pressure in the Multi-ethnic Study of Atherosclerosis (MESA). Environ Health Perspect
7.Zanobetti A, Canner MJ, Stone PH, et al. Ambient pollution and blood pressure in cardiac rehabilitation patients. Circulation
8.van den Hooven EH, de Kluizenaar Y, Pierik FH, et al. Air pollution, blood pressure, and the risk of hypertensive complications during pregnancy: the Generation R study. Hypertension
9.Kaaja RJ, Greer IA. Manifestations of chronic disease during pregnancy. JAMA
10.Ritz B, Wilhelm M. Ambient air pollution and adverse birth outcomes: methodologic issues in an emerging field. Basic Clin Pharmacol Toxicol
11.Slama R, Thiebaugeorges O, Goua V, et al. Maternal personal exposure to airborne benzene and intrauterine growth. Environ Health Perspect
12.Slama R, Darrow L, Parker J, et al. Meeting report: atmospheric pollution and human reproduction. Environ Health Perspect
13.Alperovitch A, Lacombe JM, Hanon O, et al. Relationship between blood pressure and outdoor temperature in a large sample of elderly individuals: the Three-City study. Arch Intern Med
14.Barnett AG, Sans S, Salomaa V, Kuulasmaa K, Dobson AJ. The effect of temperature on systolic blood pressure. Blood Press Monit
15.Drouillet P, Kaminski M, Lauzon-Guillain B, et al. Association between maternal seafood consumption before pregnancy and fetal growth: evidence for an association in overweight women. The EDEN mother-child cohort. Paediatr Perinat Epidemiol
16.Yazbeck C, Thiebaugeorges O, Moreau T, et al. Maternal blood lead levels and the risk of pregnancy-induced hypertension: the EDEN cohort study. Environ Health Perspect
17.Hampel R, Breitner S, Ruckerl R, et al. Air temperature and inflammatory and coagulation responses in patients with coronary or pulmonary diseases. Occup Environ Med
18.Wolf K, Schneider A, Breitner S, et al. Air temperature and the occurrence of myocardial infarction in Augsburg, Germany. Circulation
19.Cressie N. Statistics for Spatial Data. 1991
. New York: Wiley and Sons; 1991.
20.Berglind N, Bellander T, Forastiere F, et al. Ambient air pollution and daily mortality among survivors of myocardial infarction in five European cities. Epidemiology
21.Goodwin J, Pearce VR, Taylor RS, Read KLQ, Powers SJ. Seasonal cold and circadian changes in blood pressure and physical activity in young and elderly people. Age Ageing
22.Halonen JI, Zanobetti A, Sparrow D, Vokonas PS, Schwartz J. Relationship between outdoor temperature and blood pressure. Occup Environ Med
23.Hanna JM. Climate, altitude, and blood pressure. Hum Biol
24.Mordukhovich I, Wilker E, Suh H, et al. Black carbon exposure, oxidative stress genes, and blood pressure in a repeated-measures study. Environ Health Perspect
25.Ebelt ST, Wilson WE, Brauer M. Exposure to ambient and nonambient components of particulate matter: a comparison of health effects. Epidemiology
26.Brook RD, Bard RL, Burnett RT, et al. Differences in blood pressure and vascular responses associated with ambient fine particulate matter exposures measured at the personal versus community level. Occup Environ Med
27.Thornburg KL, Jacobson SL, Giraud GD, Morton MJ. Hemodynamic changes in pregnancy. Semin Perinatol
28.Peretz A, Sullivan JH, Leotta DF, et al. Diesel exhaust inhalation elicits acute vasoconstriction in vivo. Environ Health Perspect
29.Seaton A, Dennekamp M. Hypothesis: ill health associated with low concentrations of nitrogen dioxide - an effect of ultrafine particles? Thorax
30.Cheng TJ, Hwang JS, Wang PY, et al. Effects of concentrated ambient particles on heart rate and blood pressure in pulmonary hypertensive rats. Environ Health Perspect
31.Zareba W, Couderc JP, Oberdorster G, et al. ECG parameters and exposure to carbon ultrafine particles in young healthy subjects. Inhal Toxicol
32.Warland J, McCutcheon H. Is there an association between maternal hypotension and poor pregnancy outcome? a review of contemporary literature. Aust J Midwifery
33.Zhang J, Klebanoff MA. Low blood pressure during pregnancy and poor perinatal outcomes: an obstetric paradox. Am J Epidemiol
34.Xiong X, Mayes D, Demianczuk N, et al. Impact of pregnancy-induced hypertension on fetal growth. Am J Obstet Gynecol
35.Zhang J, Villar J, Sun W, et al. Blood pressure dynamics during pregnancy and spontaneous preterm birth. Am J Obstetr Gynecol
Supplemental Digital Content
© 2011 Lippincott Williams & Wilkins, Inc.