Study of the Impact of Air Pollution on Convulsions in Children: A Hospital-Based Retrospective Study in Hangzhou, China

Background: To investigate the relationship between the occurrence of convulsions in children and air pollution in Hangzhou. Methods: From January 1, 2018, to December 31, 2020, 775 children admitted with convulsion to the pediatric outpatient clinic of The Affiliated Hospital of Hangzhou Normal University (Hangzhou, China) were enrolled in this study. The clinical data and the corresponding weather data of the day in Hangzhou were collected and analyzed. Also, the monthly etiological classification of convulsions and the monthly average air data of Hangzhou were statistically analyzed. Results: The highest incidence of convulsion was observed in children 1 to 2 years old, and higher in boys than in girls. The top three main causes were febrile seizure, benign infantile convulsion with mild gastroenteritis, and epilepsy. Among the meteorological factors, the increase in the level of 2.5 micron particulate matter (PM 2.5) in the air per month led to an increase in the number of patients with febrile seizure, benign infantile convulsion with mild gastroenteritis, and epilepsy, where the increase of ozone in 8 hours (O3-8h) per month led to a decrease in the number of patients with such conditions. Conclusions: PM2.5, PM10, and SO2 are the main meteorological factors affecting the occurrence of convulsions in children in Hangzhou, and PM2.5 and SO2 are risk factors. The increase in the level of PM2.5 in the air per month could increase occurrence of child convulsions, but the increase of O3-8h per month could decrease occurrence of child convulsions.


Introduction
In recent years, more and more studies have focused on the meteorological factors influencing respiratory tract infection, cardiovascular diseases and nervous system diseases, and their correlations have been demonstrated. 1 Convulsions in children have been among the most common critical diseases in pediatric emergency departments, and the incidences of pediatric convulsions have continued to increase in recent years. 2,3 Air pollution exposure in children has been shown to increase the risk of febrile convulsions. 4 However, few studies on the correlation between the meteorological factors and convulsions in children have been performed. With the intellectualization development of meteorology, it is worth exploring how to predict the probability of children having convulsions according to changes in meteorological factors. Therefore, the purpose of this study was to study the possible correlation between the occurrence of convulsions in children and the meteorological factors in Hangzhou.

Objects
A total of 775 children with convulsion admitted to the pediatric outpatient clinic of the Hospital from January 1, 2018 to December 3, 2020, were enrolled in this study and their clinical data were collected. In addition, the meteorological data of Hangzhou (China) from the same time period were also collected, including air quality index (AQI), air pollutants, average temperature, minimum temperature, maximum temperature, air humidity, wind speed, and precipitation. The daily and monthly average air data and febrile seizure (FS), benign infantile convulsion with mild gastroenteritis (BICE), and epilepsy occurrence data of the pediatric outpatient were processed and their potential correlations were analyzed. Informed consent was obtained from all the patients' guardians before the start of the study. This study was approved by the Medical Ethics Committee of The Affiliated Hospital of Hangzhou Normal University (approval number: 2021-(E2)-HS-059).

Geographical Location of the Study City
Hangzhou is located in the northern part of the South East Coast of China. Its geographical coordinate is 118°21'-120°30'E, 29°11'-30°33'N. It is in the subtropical monsoon region, featuring abundant precipitation and four distinct seasons.

Analysis of the Clinical and Meteorological Data
The collected data included the children's age, sex, time of admission to the outpatient clinic, and causes of convulsion. The meteorological data were obtained from the Hangzhou meteorological official website of Zhejiang Meteorological Bureau and Hangzhou Environmental Inspection Center, including daily AQI, average concentration of particulate matter (PM) particles sized less than 2.5 μm (PM 2.5 ), PM particles sized less than 10 μm (PM 10 ), sulfur dioxide (SO 2 ), nitrogen dioxide (NO 2 ), carbon monoxide(CO), moving average concentration of ozone in 8 hours (O 3 -8h), daily mean temperature, daily minimum temperature, daily maximum temperature, daily air humidity, wind force, and daily precipitation. The influence of the meteorological factors on the number of children with convulsion and the total number of children admitted to the outpatient clinic was analyzed. The monthly average concentrations of the air indexes were calculated, and their correlations to the number of children with FS, BICE, and epilepsy were analyzed.

Clinical Diagnosis of Seizures
Epileptic seizure is a common form of convulsion, which is mainly characterized by the motor seizure of skeletal muscles, such as tonic and clonic seizures, and is often accompanied by consciousness disorder. 5 FS is defined as a seizure in children between 6 months and 5 years of age, accompanied by elevated body temperature (38℃ or higher) without central nervous system infection. 6 The diagnostic criteria of BICE are as follows: (1) healthy in the past; (2) aged between 6 months and 3 years; (3) mild gastroenteritis with convulsion without fever, which can be a generalized tonic-clonic seizure or partial seizure; (4) maybe accompanied by mild dehydration but no electrolyte disturbance, acidosis or hypoglycemia, and so on; and (5) normal electroencephalography, cerebrospinal fluid routine test, and cranial imaging examination during the seizure. The diagnosis of epilepsy in children was in accordance with the diagnostic criteria in the 2012 The National Institute for Health and Care Excellence (NICE) guidelines (NICE 2012). 7 If the diagnosis could not be clearly made through the clinic but was consistent with convulsion, the patients were included in the group "other."

Statistical Analysis
The SPSS version 26.0 software (IBM Corporation, Armonk, NY, USA) was used for all the statistical analyses. Continuous variables are ex-pressed as the mean ± standard deviation (χ ‾±s), and categorical variables are expressed as percentage ratio (%). We used the independent sample t test for group comparisons. For all the indexes, we used single-factor analysis and then used multiple-factor logistic regression analysis for the factors with statistical significance in the single-factor analysis. Pearson correlation analysis was used to evaluate the correlation between meteorological factors with FS, BICE, epilepsy, and convulsions in children. Pearson correlation coefficient (R) was used to express the correlation between variables. The correlation coefficient assumes values in the range from −1 to +1, where ±1 indicates the strongest possible agreement and 0 the strongest possible disagreement. The statistical significance was set as P < .05.

Clinical Characteristics and Results
Among the 775 children with convulsion, there were 446 boys and 329 girls, with a male to female ratio of 1.36:1. The top three ranges of seizure age were 1 to 2 years old (210 cases, 27.10%), 2 to 3 years old (199 cases, 25.68%), and 3 to 4 years old (105 cases, 13.55%). The number of children with convulsion was 340, 310, and 125 in the year 2018, 2019, and 2020, respectively.

Analysis of Daily Meteorological Factors and Convulsive Seizures
Single factor analysis. The dates of admission were classified into two groups according to whether there were children with convulsion admitted to the clinic or not on each day. Single factor analysis was performed on the data of the two groups and the following indexes: AQI, PM 2.5 , PM 10 , SO 2 , NO 2 , CO, O 3 -8h moving mean concentration, daily mean temperature, daily minimum temperature, daily maximum temperature, daily air humidity, wind force, and precipitation. The results revealed that the number of children with convulsion admitted to the pediatric outpatient clinic has a statistically significant correlation (P < .05) with the daily mean PM 2.5 concentration, daily mean PM 10 concentration, daily mean SO 2 concentration, daily mean NO 2 concentration, daily CO concentration, O 3 -8h moving mean concentration, mean temperature, minimum temperature, and maximum temperature (Table 1).
Correlation factor analysis. Binary logistic regression analysis was performed for the variables with statistical significance in the single-factor analysis. The analysis revealed that the average daily concentrations of PM 2.5 and SO 2 were risk factors for whether there will be children admitted with convulsion to the pediatric outpatient clinic. The average daily concentration of PM 10 was also a statistically significant factor, but it would lower the risk ( Table 2).

Analysis Between Monthly Mean Meteorological Factors and FS
Correlation analysis. Pearson correlation analysis was used to analyze the correlation between monthly meteorological factors (AQI, PM 2.5 , PM 10 , SO 2 , NO 2 , CO, and O 3 -8h moving concentrations, monthly mean temperature, minimum temperature, maximum temperature, air humidity, wind force, and precipitation) and monthly convulsions of different etiology. The results showed that PM 2.5 concentration was positively correlated with convulsions caused by FS, BICE, epilepsy, and other causes (P < .05). Additionally, O 3 -8h moving concentrations were negatively correlated with convulsions caused by FS, BICE, epilepsy, and other causes (P < .05). In terms of monthly meteorological factors, the number of outpatient convulsions in a month increased with the increase of PM 2.5 concentration in that month, while the number of outpatient convulsions in a month decreased with the increase of O 3 -8h moving concentrations in that month (Table 3).

Discussion
With the advent of the information age, meteorological prediction is becoming more and more sophisticated. Currently, there is more and more research on meteorology and diseases. 8,9 Children's convulsions account for a large proportion of pediatric emergency department visits, which is unpredictable. 10,11 Our study investigated the correlation between weather and children's convulsions, with the aim that medical staff and their families can predict children's convulsions based on meteorological forecast.
Our study found that among various daily meteorological factors, PM 2.5 , PM 10 , and SO 2 were correlated with the occurrence of convulsion in the pediatric outpatient department on that day. The higher the PM 2.5 and SO 2 concentrations on that day, the higher the probability of increased number of patients with convulsion in the pediatric outpatient department on that day, and the higher the PM 10 concentration on that day, the lower the probability of increased number of patients with convulsion in the pediatric outpatient department on that day. PM 2.5 inhalation is closely related to respiratory tract infection in children, which has been confirmed by many studies. [12][13][14] FS is the most common convulsion disorder in children, and is closely related to respiratory tract infection, which is consistent with our study. Among all the air pollutants, the daily SO 2 concentration was the most highly correlated with the occurrence of convulsions in the pediatric outpatient clinic on the same day, and the incidence of convulsions in children increased by 18.2% when the daily SO 2 concentration increased by 1 concentration unit.
In multiple studies, exposure to high concentration of SO 2 has been shown to be a threat to human health, 15 and to have a great influence on the respiratory and cardiac vascular systems. 16 However, studies on children diseases, particularly on convulsions in children, are very scarce. Some studies have suggested that endogenous SO 2 is closely related to epilepsy seizures and neuronal apoptosis. 17,18 Since our study is retrospective, we are not able to determine how the SO 2 inhaled by these children was correlated with endogenous SO 2 , but we provide a new direction for future research.
On the other hand, daily PM 10 concentration was found to be negatively correlated with the presence or absence of convulsions in children visiting the clinic on that day. PM 10 is particulate matter with an aerodynamic diameter of 10 microns or less. It should be noted that this finding does not imply that PM 10 has a beneficial effect on children's health. In fact, in many studies, PM 10 response to the pandemic was found to have a lag of 3 to 7 days 19 , which also means that attention needs to be paid to the delayed harm of PM 10 to human health. 20 In addition, our study found that among the monthly meteorological factors, the higher the PM 2.5 concentration in a month, the more children's convulsion cases caused by various causes in that month. However, the higher O 3 concentration in a month, the fewer children's convulsion cases in that month. This finding indicates that PM 2.5 is closely related to the occurrence of convulsions in children, which has a significant impact on both the current day and the current month. How to effectively reduce the probability of children having convulsions, in both the short and long term, is through a preventive effect by reducing exposure to PM 2.5 .
Furthermore, our study found that O 3 -8h was negatively associated with the admission of epilepsy patients, which had not been previously reported. Ozone exposure has been shown to be associated with seizures in an animal model, 21,22 while in child-related studies, we found only one study from Korea suggesting an association between ozone and febrile seizures. 21 Remarkably, in the current treatment of COVID-19, ozone therapy is proposed to be effective, ozone therapy can be used to control inflammation, stimulate immunity, and antiviral activity. 23,24 This may explain why in our study, O 3 -8h per month could reduce occurrence of child seizures. However, it is not known whether ozone is directly related to seizures or whether it works by controlling infections. It should be noted that excessive ozone inhalation may harm human health to some extent. High surface ozone concentration is harmful to the human body, especially to the eyes, respiratory tract, and nervous system. Surface ozone is also harmful to crops and forests. 25,26 Our study still has many shortcomings. First, it is a retrospective study with regional and ethnic limitations that only lasted for 3 years. In addition, as China has been under strict epidemic prevention and control measures since 2020, the disease spectrum related to pediatrics is changing, and patients cannot seek medical treatment due to these epidemic prevention and control measures.

Conclusions
In summary, the daily PM 2.5 , PM 10 , and SO 2 concentrations in the Hangzhou area were most closely related to the number of children with convulsions admitted to the pediatric outpatient clinic. Among the monthly mean meteorological factors in the Hangzhou area, PM 2.5 and O 3 were related to the number of children with convulsions admitted to the pediatric outpatient clinic.

Author Contributions
LY: Contributed to conception and design; contributed to analysis; drafted manuscript; critically revised manuscript. LXS: Contributed to conception; contributed to acquisition; drafted manuscript; critically revised manuscript. WC: Contributed to conception; contributed to interpretation; drafted manuscript; critically revised manuscript. YFZ: Contributed to conception, contributed to acquisition and analysis, drafted manuscript, critically revised manuscript. GSW: Contributed to conception Abbreviations: AQI, air quality index; BICE, benign infantile convulsion with mild gastroenteritis; CO, carbon monoxide; FS, febrile seizure; NO 2 , nitrogen dioxide; O 3 -8h, moving average concentration of ozone in 8 hours; PM 2.5 , particles sized less than 2.5 μm; PM 10 , particles sized less than 10 μm; SO 2 , sulfur dioxide.