Effects Of Environmental Factors And Atmospheric Pollution In The Incidence Of Hospital Admissions By Stroke In São Paulo City

Objectives: To analyze the association between ambient temperature, air humidity and atmospheric pollution with the incidence of hospital admissions for stroke in the city of São Paulo in 2016. Results: An average in the year 2016 was observed for CO of 0.58 ppm (SD = 0.12), not exceeding 0.81 ppm. For MP10, there was an annual average of 28, 26 μg / m3 (SD = 7.13), ranging from 20, 45 μg / m3 to 42.81 μg / m3. MP2.5 had a mean of 2016 of 13.40 μg / m3 (SD = 5.01). SO2 had an annual average of 1,90 μg / m3 (SD = 0,90), with a minimum of 1,20 μg / m3 and a maximum of 4,20 μg / m3. It is observed that the higher incidence values of stroke were stated in the colder and drier months at the year between May and August, that is evident looking for the female incidence of 769 in August and the male incidence of 8.92 in July. This fact matches with high concentrations of the pollutants, mostly SO2 that had the most relevant value of 4.2 μg / m3 in July.


Introduction
Stroke is one of the major causes of worldwide deaths, di cult to treat and recover from patients 1 . In more severe cases this illness can lead to physical and mental disabilities, which harms the lifestyle of many individuals 2 .
Besides, stroke damage may continue even during treatment, because of the encephalic white matter degeneration 3 . This creates an obstacle for rehabilitation, which can be an important factor for people in nancial di culties, on account of its high costs or even labor problems 4 .
In this context, the study of some environmental factors, which leads to an increase of stroke incidence, is important because the treatment and prevention of this morbidity are more effective in a risk control scenario 5 . Moreover, the treatment methods of stroke are precarious in many Brazilian areas, uctuating a lot even in the same town 6 .
Environmental factors have great impact on population heath, being that air pollution was associated with at least 5.708.000 deaths all over the world in 2015 7 . These characteristics surround all the individuals and affects the type of work produced by each person daily, so the quality of life is closely related to the ambient 8 .
The atmospheric pollution, temperature and air humidity, already have been associated with stroke risk [9][10][11] . This can corroborate the increase in cases of stroke in those regions 12 . Therefore, despite the fact that stroke incidence has been falling in Brazilian Southeast in the period from 2008 to 2012 13 , the numbers of death by this disease is increasing due to population rises and epidemiology transition in the country 14 .
The residing population in São Paulo, a global metropole is in uenced by air pollution and climate changes, due to the accumulation of pollutants in this place by vehicular tra c and industrial gases emitted in the region 15 Stroke, as already seen, has high rates of mortality and high incidence worldwide 16 , as in Brazil. The relationship of this disease with pollution and temperature has already been considered in other studies 17,19 . However, in 2016, the city of São Paulo experienced high pollutant rates in addition to very high or very cold temperatures for national standards. This has raised doubts about how these environmental variables can affect the incidence of hospital admissions for stroke in a large global metropolis.
Thus, the main objective of this article is to analyze the association between ambient temperature, air humidity and atmospheric pollution with the incidence of Stroke in the city of São Paulo, Brazil, in 2016.

DESIGN OF STUDY AND POPULATION
This is an observational, Ecological study that uses secondary data referring to hospital admissions for Stroke and environmental conditions of São Paulo city, Brazil, in 2016.
The study was conducted with individuals living in the capital of São Paulo, the city with the largest population accumulation in the country, according to the Census of Brazilian Institute of Geography and Statistics 18 .

STUDY VARIABLES
The variables of interest that were studied were: 1. the incidence of hospital admissions for stroke in São Paulo 2. the monthly mean temperature, humidity and air pollutants (particulate matter and pollutant gases) in the city of São Paulo.

DATA COLLECT
The de nition of stroke is in accordance with the tenth revision of the International Classi cation of Diseases (ICD10) codes: I60, I61, I63, and I64, which correspond to intracranial hemorrhage, cerebral infarction, stroke not speci ed as hemorrhagic or ischemic in the morbidity list of ICD-10. The number of hospital admissions in the city of São Paulo was obtained by the Hospital Information System of the Uni ed Health System (SIH / SUS), which are available in DATASUS. This database is the o cial health information of the Brazilian Ministry of Health, and is available for free access to the user at http://datasus.saude.gov.br/.
CETESB is the agency of the state of São Paulo responsible for the control and monitoring of environmental conditions in the region. Data of this company were used to analyze the temperature of the environment, the humidity of the air and the concentration of particulate material and pollutant gases in the place and period studied is this data available in the site http://www.cetesb.sp.gov.br/.

STATISTICAL ANALYSIS
For the descriptive analysis of the quantitative variables, as they had a normal distribution (Shapiro-Wilk test, p > 0.05), mean, standard deviation, minimum and maximum were used.
To analyze the association between exposure to pollutants and the incidence of hospital admissions for stroke, the Pearson and Spearman correlation test will be used. The con dence level adopted will be 95% and the statistical program used will be the Data Analysis and Statistical Software for Professionals (Stata) version 13.0®.

ETHICS COMMITTEE
As this study is based on secondary data, since it is not able to identify the individual and the data is free and unrestricted on the Internet, there is no need for this project to be sent to the Research Ethics Committee for its appreciation, as expressed in the resolution 466/2012.  (Table 1).

Discussion
The current study was carried out in the city of São Paulo, a large, heavily industrialized state region. In this study, the main ndings were the higher rate of hospital admissions for stroke in males compared to  In this study, it is observed that the higher incidence values of stroke were stated in the colder and drier months at the year between May and August, that is evident looking for the female incidence of 769 in August and the male incidence of 8.92 in July. This fact matches with high concentrations of the pollutants, mostly SO2 that had the most relevant value of 4.2 µg / m3 in July (Fig. 1). There was a negative correlation between temperature and the incidence of total stroke in women and men. In the case of relative air humidity, there was a positive correlation with the incidence of total stroke for men, however, this correlation is negative for women. It is also observed that only correlations involving SO2 were statistically signi cant (Table 2).
females, in addition to the positive correlation found between the concentration of SO2 and the incidence of stroke.
One study analyzed mortality from stroke in the large Brazilian metropolises, including São Paulo, and found that mortality in men due to this morbidity is higher when compared to women 20 . This may be related to the higher incidence of male hospital admissions found in the present study, as the samples in both cases are similar.
In addition, the higher rate of hospital admissions in men compared to women has already been described in a review proposed by Piassaroli et al, who used a total of 92 articles on the subject related to stroke 21 , and most of these articles proves a similar scenario, with a higher incidence in men. Furthermore, in an ecological study similar to this, Alcantara et al, found that the incidence of stroke in men is 19% higher than in women 22 . Thus, the present study follows the same proportion, with the incidence in men being 7.61 and in women being 6.38, which shows that our results corroborate a tendency regarding sex.
The positive correlation between the SO2 concentration and the incidence of hospital admissions for stroke in both sexes may be justi ed by the absorption of this gas in the membranes of the respiratory tract, which may contribute to the activation of neuronal signals that stimulate bronchial constriction and cardiovascular, as already explained by Tunnicliffe, in a study on the effects of SO2 in adults, where he found that this gas participates in neural mechanisms 23 .
In this scenario, it is possible to notice that the increase of SO2, generates an increase in hospital admissions for stroke in São Paulo, which are more present in the colder months of the year, periods in which the SO2 rates reach the highest averages. This increase in SO2 in these months can be understood as a consequence of the thermal inversion, which occurs in the coldest time of the year, reducing the dispersion of pollutants 17 .
Other pollutants studied were positively correlated with the incidence of stroke. However, no other pollutant in question had a statistically signi cant result (p > 0.05). Huang et al. stated in their study that it is not possible to accurately correlate stroke with particulate matter, which corroborates our result, since this correlation can vary even in the same year in different seasons 1 . In the case of carbon monoxide, the statistically insigni cant result may be in uenced by a protective role that this gas promotes in low concentrations against ischemic stroke, as already explained by Wang et al. 24 in his study on the effects of CO as a neuro protector.

Conclusion
Stroke is, indeed, a morbidity that affects many individuals in the city of São Paulo. This impairs public nances, the di cult treatment, which involves long periods of rehabilitation, or even the dependence on the care of the patients of stroke for the rest of their lives, being an important negative factor for the quality of life of the population, which has accessible information to prevent this disease.
In the present study, it was observed that there is a positive correlation between total hospital admissions for stroke and the concentration of atmospheric pollutant Sulfur Dioxide (SO2). This fact must be considered in the preparation of public measures for the intervention in the dispersion of pollutants, so concentrated in the city of São Paulo.

Limitations
The faint association of other pollutants, but SO2, can be considered a negative point of this study. This can be explained by analyzing each pollutant separately and in a short period, which can camou age possible reactions between the organism and all pollutants together. In a review, for example, by Ljungman et al. 25 , it was possible to point out that such correlation has already been studied in developed countries. In addition, Lipsett et al 26 stated that long-term exposure to particulate matter and ne particulate material signi cantly increases the risk of stroke. How this is a study that uses secondary data, it is not possible to identify the subjects of the research and because it has free and unrestricted access on the Internet, it is exempt from the evaluation of the research ethics committee according to resolution 466/2012 of the national health council.

Consent to publish
That the manuscript has been approved by all authors, who agree to cede the copyrights to the Journal Availability of Data and Materials Figure 1 Graphic illustration of the atmospheric pollution variation by month and its correlation with the incidence of Stroke between a year.