High-normal blood pressure (prehypertension) is associated with PM2.5 exposure in young adults

We aimed to examine PM2.5 exposure, blood pressure (SBP and DBP) measurement, and hypertension risk factors and to assess the association between PM2.5 exposure and hypertension among young adults. The mean SBP was 117.78 mmHg, with 11.22% high-normal blood pressure (prehypertension) and 2.51% hypertension (≥ 140 mmHg). DBP was 75.48 mmHg with 26.37% prehypertension and 4.53% hypertension (≥ 90 mmHg). The median PM2.5 in the past year was 31.79 μg/m3, with highest in winter (49.33 μg/m3), followed by spring (37.34 μg/m3), autumn (29.64 μg/m3), and summer (24.33 μg/m3). Blood pressure was positively correlated with age, height, weight, BMI, daily smoking, alcohol consumption, mental stress, and staying up in the past 1 year, and negatively with season-specific temperature. After adjustment for the covariates, each 10 μg/m3 increase in PM2.5 was associated with SBP (day 1 = 1.07 mmHg, day 3 = 1.25 mmHg, day 5 = 1.01 mmHg) and DBP (day 1 = 1.06 mmHg, day 3 = 1.28 mmHg, day 5 = 1.29 mmHg, day 15 = 0.87 mmHg, day 30 = 0.56 mmHg). Exposure in winter and the past year was associated with 1.21 mmHg and 0.95 increase mmHg in SBP, respectively. Logistic models showed for every 1 μg/m3 increase of PM2.5, SBP in day 1 and day 5 was increased by 6% and 4%, and DPB by 3% and 16%, respectively. SBP was increased by 8% in spring and 19% in winter, and DBP was increased by 7% in winter. Our data suggest a certain prevalence of pre- or hypertension among young population, which is associated with short-term fluctuation and season-specific exposure of PM2.5.


Introduction
Cardiovascular disease is the leading cause of deaths worldwide. It is estimated that 17.8 million people die for it globally in 2017, among which hypertension is the main risk factor for cardiovascular disease and is considered the most important cause of disability and death worldwide . A recent study finds that compared with data in 1958 ~ 1959, 1979 ~ 1980, 1991, 2002, and 2012, the crude prevalence rate of hypertension among Chinese residents aged 18 years and above ranges from 18.0 to 44.7% between 2012 and 2015, and the overall prevalence is increasing (Yin et al. 2021). China Health and Nutrition Survey shows that teenagers aged from 6 to 17 years old suffer an increased risk of hypertension from 7.1% in 1991 to 13.8% in 2009, and 6.4% of school-age children have abnormally high blood pressure, and a considerable part of them will develop hypertension in adulthood (Wang et al. 2019b). These results indicate that the incidence of hypertension is gradually becoming younger and has become an important public health problem.
The pathogenesis of hypertension is complex and has many influencing factors. In addition to genetics, diet, and lifestyle, changes in blood pressure are also related to air pollution (Braziene et al. 2020;Zeng et al. 2021). Air pollutant-associated health effects have long been concentrated and studied. It can promote or induce the development of hypertension through imbalance of the autonomic nervous system, oxidative stress, immune inflammation, and endothelial dysfunction (Zhao et al. 2019a). In recent years, epidemiological studies have reported that air pollution, especially PM 2.5 , may be one of the pathogenic factors of hypertension, and is related to the incidence and mortality of cardiovascular diseases . However, the differences between the levels of PM 2.5 and the research design in different regions make the conclusion inconsistent (Rabito et al. 2020). This study aims to investigate PM 2.5 level at a fixed-point monitoring site, blood pressure changes in young adults and the related risk factors, to confirm the relationship between blood pressure and PM 2.5 exposure by adjusting individual lifestyle and behavior.

Study population
As undergraduates are from different department and classes, a cluster stratified sampling method was adopted to recruit the participants of different ages between October 2019 and November 2019 in Jiaxing University, Jiaxing City, China. The sampling area (30°77′ N, 120°76′ E) is located in the northern part of Zhejiang province. Those who lived in the local area for more than 1 year, with no family history of hypertension, pulmonary, cardiovascular, and other chronic diseases were included in the study. This study was approved by the institutional review boards of all participating institutions including the Human Ethical Committee of Jiaxing University Medical College (JUMC-IRB-2018).

Questionnaires
An initial number of 900 questionnaires were delivered and the final collected number was 838 (93%). The questionnaire was designed according to the research purpose, domestic and foreign related literature, and the actual situation of the preliminary survey. The reliability and validity of the preliminary questionnaire were tested in order to improve the existing problems. The content of the questionnaire involves three aspects: (i) demographic information including gender, birth date, family income per year, residence time living in local, family history of hypertension, pulmonary, cardiovascular, diabetes, and other acute or chronic diseases; (ii) including cognitive behavior of hypertension including diagnostic criteria (1 = 140/90 mmHg; 2 = 160/95 mmHg; 3 = 130/80 mmHg; 4 = Unknown; 140/90 mmHg as the correct diagnostic criteria for hypertension), complications (1 = Stroke/coronary heart disease/tumor; 2 = Unknown) and treatment of hypertension (1 = Medicine treatment only; 2 = Nonmedical treatment (including keeping good mood, proper diet and exercise according to doctor's advice); 3 = Both; 4 = Unknown; the correct treatment for hypertension referring to "Both"), checking blood pressure regularly at home or in the hospital (1 = At least once a week; 2 = Once in a quarter; 3 = At least half a year; 4 = Uncertainty), etc.; (iii) lifestyle and activities such as intake of salty foods, smoking, excessive drinking, exercise, mental stress, and staying up.

Blood pressure measurement
The blood pressure measurement including systolic blood pressure (SBP) and diastolic blood pressure (DBP) was scheduled from 08:00 to 10:00 every morning, which was performed under the trained professional researchers. Before the measurement, participants are required to sit still for at least 5 min. The blood pressure of the right brachial artery was measured via a medical-grade arm digital automatic sphygmomanometer BA-806 with a testing error of 3 mmHg (± 0.4 kpa) (Panasonic, Japan). The measurement is repeated for 3 times each. The interval between measurements was at least half a minute. Clinical hypertension standard was set at ≥ 140/90 mmHg, and high-normal blood pressure (prehypertension) at 130-139/80-89 mmHg (Lurbe et al. 2019).

PM 2.5 data monitoring
The sampling site of Jiaxing University was one of the environmental monitoring sites in Jiaxing City (Fig. 1), and the average distance from the ground monitor of air quality to locations of participants was about 0.20 km. Air pollution indicators including PM 2.5 exposure were measured every day. We collected PM 2.5 data from the National Environmental Monitoring Center (http:// www. cnemc. cn/). In this study, the length of exposure for participants was divided into short-term exposure based on the exposure accumulation in the past 1 month (days 1, 3, 5, 7, 15, 30) (Li et al. 2016;Wellenius et al. 2013), and relative long-term exposure in the past 1 year (spring, summer, autumn, winter) (Tan et al. 2018). In detail, day 1 pointed to the median level of the past day, and day 3 was the median concentration during the past 3 days, and other concentrations were similarly calculated.

Statistical analysis
Microsoft Excel and SPSS22.0 software were applied to manage and analyze the data. Measurement data were expressed as mean ± standard deviation (SD), and categorical data were described as the number of cases (%). PM 2.5 level was skew distribution and presented as median ± SE (min ~ max), and the comparison between multiple groups was performed by the Kruskal-Wallis H test with a further comparison by post hoc test. Cognitive and behavioral factors related to hypertension were multi-class data, and the comparisons between groups were analyzed by χ 2 test. Pearson and Spearman correlations were used to analyze the risk factors of hypertension. Generalized linear mixed model was further used to analyze the relationship between blood pressure and PM 2.5 levels (Ren et al. 2019). In the model, SBP and DBP were respectively used as dependent variables, PM 2.5 normally distributed after logarithmic transformation was as factors, and the observed individual number was used as the main variable in the random effect model, with statistically significant risk factors as covariates (confounding factors) to be adjusted. Factors and covariates were included in the main effects of the fixed model, with results of the model effect test to determine whether it was statistically significant (β; 95% confidence). β coefficient indicated a unit change in blood pressure caused by every 10 μg/m 3 increase of PM 2.5 exposure. The binary logistic regression model was applied to analyze the relationship between PM 2.5 exposure and the risk of hypertension. Because of the quite limited number of hypertension subjects, the model used a binary classification of high-normal blood pressure (SBP ≥ 130 mmHg or DBP ≥ 80 mmHg) as the dependent variable, and short-term or long-term PM 2.5 as a covariate with adjustment for confounding factors (Curto et al. 2019). OR value represented the risk of a unit change in blood pressure for an increase of 1 μg/m 3 of PM 2.5 concentration. A p value less than 0.05 or 0.01 in a two-tailed test was considered to be statistically significant.

Characteristics of the study population
The general characteristics are shown in Table 1. The average age of subjects was 19.19 ± 1.35 (years), and there were 327 male (39.02%) and 511 female (60.98%). The average systolic blood pressure (SBP) was 117.78 mmHg with high-normal blood pressure accounting for 11.22%, and ≥ 140 mmHg accounting for 2.51%. The diastolic blood pressure (DBP) was 75.48 with high-normal blood pressure accounting for 26.37%, and ≥ 90 mmHg accounting for 4.53%. There were 354 participants (42.24%) having a family history of hypertension.

Cognition and behavior of hypertension
There were statistical differences in the awareness rate of hypertension (p < 0.001) ( Table 2). Further analysis found that the correct rate of the diagnostic criteria for Fig. 1 The geographic distribution of sampling area in this study hypertension was the highest (41.41%), and the correct rate of hypertension treatment accounted for 90.33%. The behavioral factors such as eating greasy food, smoking daily, and excessive drinking were all statistically different (p < 0.001).

Potential factors related with blood pressure
Pearson and Spearman correlation analyses showed that both SBP and DBP were positively associated with age, height, weight, BMI, smoking every day, staying up, and long-term temperature (Table 3). Smoking every day, excessive alcohol consumption, and mental stress in the past year had weak association with SBP and DBP. Gender showed an opposite association with SBP and DBP, indicating gender as an influencing factor. DBP but not SBP displayed a negative weak correlation with temperature in summer and autumn, suggesting an impact of season-specific temperature. Shortterm temperature showed no statistical correlation (data not shown).

Risk of hypertension
Binary logistic regression models for the association of short-term or long-term exposure with SBP and DBP were assessed, in which the blood pressure was classified into normal blood pressure and high-normal blood pressure as the dependent variable. The short-term models were adjusted for age, gender, height, weight, BMI, eating greasy food and salty food, every day smoking, excessive alcohol consumption, and mental stress in the past year, while longterm models adjusted temperature in addition to the above covariates (Table 4). Increments of 1.06 mmHg (95%CI: 1.01, 2.11) and 1.04 mmHg (95%CI: 0.97, 1.71) in SBP, and 1.03 (95%CI: 1.01, 1.79) and 1.16 (95%CI: 1.02, 2.01) in DBP were positively associated with short-term PM 2.5 exposure at day 1 and day 5, respectively. PM 2.5 exposure in spring and winter increased the risk of SBP per unit increase of 1.08 mmHg (95%CI: 0.57, 1.62) and 1.19 mmHg (95%CI: 0.37, 2.04), respectively. PM 2.5 exposure in winter increased the risk of DBP per unit increase of 1.07 mmHg (95%CI: 0.25, 1.92).   (Zhao et al. 2019b). This indicates that there are still certain problems in the air quality in local. Though overall PM 2.5 in long term is low, the exposure level exceeds the national standard limit within days, and season-specific especially for winter should be alerted. Therefore, the monitoring in this place should be continuously strengthened. Hypertension is one of the risk factors for cardiovascular disease. There is 23.2% (≈ 244.5 million) of the Chinese adult population ≥ 18 years of age having hypertension, and 41.3% (≈ 435.3 million) having high-normal blood pressure (prehypertension) according to the Chinese guideline ). Data in this study found that the average SBP was 117.78 mmHg, and the prehypertension accounted for 11.22%, and hypertension (≥ 140 mmHg) for 2.51%. The average DBP level was 75.48 mmHg with prehypertension 26.37%, and hypertension (≥ 140 mmHg) 4.53%. The prevalence of prehypertension and hypertension is lower than that in other studies according to a prior investigation in 2017 in China showing that the hypertension prevalence ≥ 18 years among 14, 220 permanent residents is up to 46.9% (Shen et al. 2017). A recent study from a cohort of 20-year followup in China manifests that compared with the young and middle-aged population with no cardiovascular disease at baseline BP or lower than 120/80 mmHg, the risk of cardiovascular disease including coronary atherosclerotic heart disease (coronary heart disease), stroke, and cardiovascular death is significantly increased in the pre-hypertensive stage at 130-139/80-89 mmHg, and more than 60% of the pre-hypertensive middle-aged and young people have blood pressure progressed to ≥ 140/90 mmHg after 15 years, with  their cardiovascular disease risk three times higher than that of normal BP (Qi et al. 2018). In addition, we found that the accuracy of response to the diagnostic criteria of hypertension among young individuals was no more than a half (41.41%). Among the behavioral factors, BP measurement at least once a week accounted for only 4.06%, but quite a few liked to eat oily food (25.54%) and salty food (35.08%). A small number of young students (17.18%) had mental stress in the past year, and most students stayed up late (versus 7.76% did not stay up). These results suggest that young adults are under unreasonable diet and lifestyle with little awareness of the risk of hypertension. Health promotion and education of hypertension for young people should be paid attention, and those with high-normal blood pressure should receive clinical check.
In the study, SBP and DBP to some extent were both positively associated with age, height, weight, BMI, smoking every day, excessive alcohol consumption, mental stress in the past year, and staying up. Though temperature has been considered an important factor contributing to BP risk, in the present study, temperature in summer and autumn showed weak inverse associations with DBP but not SBP, suggesting Fig. 3 Estimating of blood pressure changes per 10 μg/m 3 increase of PM 2.5 exposure by generalized linear mixed model analyses (β; 95%CI). Short-term models were adjusted for age, gender, height, weight, BMI, smoking every day, eating oily food, excessive alcohol consumption, mental stress, and staying up. Long-term models were adjusted for the above covariates as well as the average temperature an impact of seasonal temperature change on BP. This phenomenon is similar with a recent study that diurnal temperature range presents an opposite effect on BP, with a positive linear correlation with SBP while a negative with DBP (Zheng et al. 2020). Studies have shown that the occurrence and development of hypertension are closely related to age, overweight and obesity, family history of hypertension, alcohol consumption, high uric acid, and high C-reactive protein (Flack and Adekola 2020). This demonstrates that diet, lifestyle, and mental state are related to changes in blood pressure, and unreasonable diet, lifestyle, and mental stress are potential risk factors for hypertension. In recent years, studies have found that PM 2.5 exposure levels are related to hypertension, but there are differences between the research conclusions. Ren et al. (2019) used a general linear model to analyze the association between daily and fixed monitoring point PM 2.5 exposure and blood pressure among young people in short-term 1 to 3 days, exhibiting that PM 2.5 exposure is related to the reduction of ambulatory blood pressure, and after adjusting for age and other confounding factors, SBP and DBP were decreased by 0.54 mmHg and 0.22 mmHg for an increase of 10 μg/m 3 PM 2.5 , respectively, while decreased by 0.95 mmHg and 0.74 mmHg at fixed monitoring stations, respectively. Baumgartner et al. (2011) analyzed the relationship between 24-h PM 2.5 exposure and blood pressure among female adults ≥ 25 years by mixed linear models, and find that an elevation of 10 μg/m 3 increase in PM 2.5 is associated with an increase of 2.2 mmHg in SBP and 0.50 mmHg in DBP. Our study found that SBP was increased by 1.07, 1.25, and 1.01 mmHg for every 10 μg/m 3 increase in PM 2.5 at day 1, day 3, and day 5, respectively, and DBP increased by 1.06, 1.28, 1.29, 0.87, and 0.56 mmHg at days 1, 3, 5, 15, and 30, respectively. For long-term exposure, an increase of 1.17 mmHg in SBP was associated with each 10 μg/m 3 increase in PM 2.5 level in winter. These results show that the PM 2.5 exposure level in this area is related to the increase in blood pressure. Further we adjusted the potential confounding factors, finding that for every 1 μg/m 3 increase in PM 2.5 , the risk of SBP was increased by 6% and 4% at day 1 and day 5, respectively, and DBP risk increased by 3% and 16%, respectively. After adjusting relevant covariates including temperature, SBP risk was increased by 8% in spring and 19% in winter, and DBP was increased by 7% in winter. Some studies have shown that by modifying blood pressure-related factors such as psychosocial stress (Hicken et al. 2014), gender and age (Baumgartner et al. 2011), the influence of short-term PM 2.5 on blood pressure is changed, suggesting that potential risk factors should be considered when assessing the exposure association with PM 2.5 . The concentration of PM 2.5 also plays a critical role as exposure in winter or spring is relatively higher than other seasons, and makes a significant change in SBP or DPB. We also tried to explore the detailed linkage between PM 2.5 exposure and blood pressure by stratification of gender or age, but the model failed to distinguish due to the limited sample size of subjects when separating into different gender or age. Thus, we adjusted the gender and age in the whole model. The mechanism of PM 2.5 exposure for blood pressure change also has been extensively studied. PM 2.5 exposure can activate the inflammatory response in the arcuate nucleus of the thalamus in the acute blood pressure response, leading to upregulations of pro-inflammatory factors and inhibitory factors kappaB kinase (IKK)/nuclear factor kappaB (NF-κB), which is related to abnormal activation of the cerebral sympathetic nervous system (Ying et al. 2014). PM 2.5 exposure can also induce inflammation and oxidative stress in the circulatory system of hyperlipidemia rats, activating the JNK/P53 pathway and promoting hypercoagulability and cardiomyocyte apoptosis (Wang et al. 2019a). Therefore, PM 2.5 exposure can affect blood pressure by activating oxidative stress or inflammation. This study also has certain limitations. First of all, the scope and time of indoor or outdoor activities of subjects are different, and the PM 2.5 monitoring at a fixed station cannot accurately reflect their daily exposure level, and there is a certain bias. Secondly, the clinical recommendation of blood pressure measurement is to maintain a resting state when you wake up in the morning, and the measured blood pressure is relatively accurate. This study selected a time period in the morning which had little effect on the stable blood pressure individuals, but it might bring some changes to those with significant fluctuating people. Air pollutant exposure is quite complex posing a wide range of health effects on the population, but we did not include all such as PM 10 , SO 2 , NO, and humidity. We only focused on the relationship between PM 2.5 exposure and blood pressure changes since the concentration of PM 2.5 pollutant is highlighted and typical in this area.

Conclusions
In summary, there is certain prevalence of prehypertension and hypertension among young college students in local. In addition to age, gender, height, weight, BMI, eating oily and salty food, smoking every day, excessive alcohol consumption, and mental stress in the past year, PM 2.5 in short-term periods or season-specific exposure is related to changes in blood pressure, which can increase the risk of prehypertension. Therefore, attention should be paid to the early prevention of prehypertension and hypertension, to effectively reduce the incidence of hypertension in the young population.