In this cross-sectional study we observed that children attending a primary school situated in an area of high air pollution, close to heavy traffic, had a substantially higher arterial blood pressure than children attending a school situated in an area with less air pollution. Our findings are novel on at least two accounts: to our knowledge, our study is the first to investigate the adverse health effects of urban air pollution in children from Pakistan and, moreover, this is one of the first studies to report an association between particulate air pollution and blood pressure in children. With mean differences of 7.6
mmHg for systolic blood pressure and 4.5
mmHg for diastolic blood pressure, the differences between the two groups of children are not only statistically significant, they are also substantial and, therefore, likely to be biologically relevant with regard to the long-term risk of cerebrovascular disease and coronary events.
The differences in blood pressure between the two groups are unlikely to be due to technical factors. Blood pressure was measured according to guidelines of the European Society of Hypertension [19
] using a technically validated automated device, thus eliminating any investigator-dependent bias [23
]. All measurements were made in similar, standardized circumstances, over a period of 4
months (January to April 2009), with the children having been sitting for at least 5 minutes, and with 5 consecutive readings being obtained.
In the absence of technical factors or measurement errors that could explain the differences in blood pressure between the two groups, the following factors can be invoked to explain why the children from the most polluted environment had a higher blood pressure: dietary differences (with a higher salt intake), higher exposure to traffic-related noise and air pollution.
The dietary intake of sodium is a well-established determinant of blood pressure in adults and children [24
]. The two groups of schoolchildren differed in their socio-economic status (as assessed by father’s educational level) and it is, therefore, conceivable that they also differed with respect to their diet, either qualitatively or quantitatively, and so we have adjusted for socio- economic status as a confounder in our analyses. However, we do not believe these differences in socio-economic status to be very important. First, the two groups of schoolchildren did not differ with regard to height and weight, thus excluding any large quantitative differences in food intake. Second, the two groups did not differ significantly with regard to their urinary sodium concentrations, thus suggesting similar intakes of salt for both groups. The values of creatinine did differ considerably between the two groups, with lower values (mean of 0.76
g/l) being found among the low pollution group and higher values (mean 1.32
g/l) among the high pollution group. We have no explanation for this finding. Technical factors do not seem to be a reason for this finding, because creatinine concentrations were reassessed in Belgium in a random 10% of the samples and found to be similar as the ones measured locally (data not shown). There was undoubtedly more noise at the school exposed to high pollution than at the less polluted school. Minimum and maximum sound levels were indeed higher in the high pollution school than in the low pollution school, but this does not mean that the average noise level was higher by 8 to 12
dB (A) in the high pollution school compared to the low pollution school, because the instrument used did not allow us to assess time-weighted average noise levels. It was, therefore, impossible to correct for any effect of exposure to noise during the measurements of blood pressure. Moreover, as suggested by various investigators [25
], to the extent that traffic causes both noise and air pollution, it is difficult to disentangle the effects of these two factors. Moreover, the results of studies on the effects of (road and aircraft) noise have been somewhat contradictory [27
]. In a recent study of 1048 German schoolchildren, children living in busy traffic streets had only slightly higher values of systolic blood pressure (1.8
0.05) and diastolic blood pressure (1.0
mmHg, NS) than those living in low traffic streets [28
In preschool children from Belgrade, a higher difference in systolic blood pressure (5
mmHg) was found in children from noisy residences compared to those from quiet residences [29
]. So, we cannot exclude that differences in exposure to traffic noise were partly responsible for the differences in blood pressure between our two groups of children.
In view of the established effects of particulate pollution on the cardiovascular system, we must consider traffic-related air pollution by particulate matter as a plausible explanation too. In studies of air pollution, it is important to distinguish short-term temporal effects of peaks of air pollution - which act as triggers of adverse events such as myocardial infarction [30
] - from the chronic effects of continuous exposure to air pollution. Although we did measure daily PM levels in our study, and although these varied somewhat from day to day during the course of the study, there were no significant relations within each group between blood pressure and current PM or PM over the past 24 hours (data not shown). These relations were, however, significant when we considered the entire group, but this was dominated by the spatial contrast in PM between the two groups, because the effects of PM remained when we took temporal variability into account in the analysis. Consequently, we attribute the difference in blood pressure between the two groups of schoolchildren to the difference in their chronic exposure to air pollution.
Many epidemiological studies have found a consistent relation between long-term exposure to urban particulate exposure and morbidity or mortality from cardiovascular disease in adults, and experimental studies provide mechanistic plausibility for these associations [3
]. Several recent studies have also found proximity to heavy traffic to be more closely associated with adverse health effects than estimated exposure to particulate matter [1
]. Very few studies have reported on cardiovascular effects of proximity to traffic in children. Iannuzzi et al. [33
] found a statistically significant difference in carotid stiffness, but not in carotid intima-media thickness or peripheral artery pressure, among 52 children according to their estimates exposure to air pollution. Calderón-Garcidueñas et al. [34
] attributed to air pollution the higher pulmonary arterial pressure found in children living in Mexico City, compared to control children exposed to lower levels of PM2.5
, but they did not report on arterial blood pressure in these children. Recently, a large epidemiologic study showed that preschool children exposed to environmental tobacco smoke had on average 1
mmHg higher systolic blood pressure than children not exposed to environmental tobacco smoke [35
]. As with polluted ambient air, environmental tobacco smoke is largely composed of an aerosol of particles derived from combustion [36
], therefore, our conclusions for outdoor air pollution and those for passive smoking mutually support each other.
What are the possible implications of our observations that children living and attending school in an environment with higher levels of air pollution, presumably as a result of traffic, have a higher arterial blood pressure than children living in a less polluted environment? Although no child had hypertension and although the mean systolic blood pressure in the most exposed group was below 140
mmHg, mean differences of about 7
mmHg for systolic blood pressure and about 4
mmHg for diastolic blood pressure are substantial. Thus, in our population, the odds of having a blood pressure above “normal” blood pressure (systolic <120
mmHg and diastolic <80
mm Hg) was 2.56 times higher in those living in the highly polluted area. Admittedly, the criteria used here for defining normal blood pressure and pre-hypertension are those applicable for adults from industrially developed countries [20
]. Although it remains to be determined whether the same limits apply for the children studied here, we have used these categories to put our results into some clinical perspective. At the population level, pre-hypertension (systolic 120–139
mmHg or diastolic 80-89
mmHg) is associated, in adults, with a higher incidence of hypertension [37
] and an increased risk of cardiovascular complications [38
] compared with normal blood pressure. Risk factors in children may persist into later life and eventually lead to cardiovascular disease [39
]. Findings from the Bogalusa Heart Study in the USA demonstrated that childhood blood pressure levels at or above the 80th
not necessarily in hypertensive ranges, were associated with an increased prevalence of elevated blood pressure during adulthood [39
]. A follow-up study [13
] showed that blood pressure in a group of male students at the age of 20.5
years was associated with the incidence of cardiovascular diseases in the following 41.3
years. These findings indicate that elevated blood pressure during young age may have later clinical significance.
In the absence of similar studies from other areas, we do not know how representative our findings are for other populations. It has to be noted that in the present study, the outdoor PM levels measured in the high pollution area were extremely high (average PM10
, average PM2.5
) and still high, by Western standards, even in the “less polluted environment” (average PM10
, average PM2.5
). These relatively low proportions (15-25%) of PM2.5
indicate that in Lahore - as in other locations from developing and emerging countries [15
] - the high PM levels are largely due to coarse particulates, i.e.
presumably resuspended dust.
Our study has limitations including the cross-sectional nature. Thus the blood pressure measurements were taken on just one day, which did not account for inter-day variability. We did not measure gases (NO2 and O3), nor residential PM. Also, personal exposure to particulate air pollution was not recorded. Due to the aggregation of location and air pollution, we could not control for location in the statistical models. Another limitation is that we only studied a limited number of children from just two schools, so we do not know how representative the results are for the whole city of Lahore, let alone other cities. However, the results of our study justify a larger study on the effect of PM on blood pressure in children.