The Ionic Component of PM2.5 May Be Associated with Respiratory Symptoms and Peak Expiratory Flow Rate

Featured Application

Few studies have evaluated the relationship between the concentration of ionic components in PM2.5 and clinical symptoms or peak expiratory flow rate. The present study is an important stepping stone for future research.

Abstract

(1) Background: Few studies have evaluated the association between the ionic components of PM2.5 and respiratory symptoms or peak expiratory flow rate (PEFR). The aim of this study was to clarify these associations. (2) Methods: Patients with allergic diseases treated at Fukuoka Hospital recorded their morning and evening PEFR and respiratory symptoms in a diary. We measured PM2.5 and its ionic components in Fukuoka City and examined the relationship with PEFR and respiratory symptoms by univariate and multivariate analysis. (3) Results: Among the ionic components of PM2.5, Cl⁻, NO₃⁻, Na⁺, K⁺, and Mg²⁺ were significantly correlated with the frequency of coughing and nasal symptoms. In univariate analysis, the concentrations of each of the above ions was significantly associated with a decrease in PEFR and the concentrations of each was associated with at least one respiratory symptom or PEFR. Multivariate analysis of items significantly correlated with PEFR indicated that the concentration of ionic components may predict changes in PEFR. (4) Conclusions: In patients with allergic diseases, some ionic components of PM2.5 may increase the frequency of respiratory symptoms and decrease PEFR, so further study and caution are required in daily clinical practice.

1. Introduction

The prevalence of asthma in Fukuoka city, Kyushu is at a relatively high level for Japan, with Kyushu having both a high prevalence of bronchial asthma (BA) and high atmospheric levels of PM2.5 and SPM [1].

There have been several reports about air pollution, allergies, and respiratory diseases in Fukuoka. A number of studies have also examined the relationship between respiratory symptoms and air pollution [2,3], especially PM2.5 (particulate matter with an aerodynamic diameter less than 2.5 μm) with similar results. [4,5].

However, reports on the health effects of PM2.5 have not always yielded consistent results. There are several reasons for this. For example, a combination of PM2.5 exposure (mainly associated with traffic-related air pollution) and secondary sulphate sources (mainly associated with industrial activity) was found to significantly exacerbate respiratory symptoms in children [6].

In this way, there are many factors that can be considered, such as the combination of ingredients that affects the effect, and the effect that differs depending on age. One of them is the difference in the ionic components of PM2.5.

In other words, it is considered necessary to examine the ionic components [7]. However, few studies have evaluated the relationship between the ion content of PM2.5 and peak expiratory flow (PEFR) or respiratory symptoms in real allergic patients.

We hypothesized that the ionic components of PM2.5 influence his PEFR and symptoms of allergic patients.

Therefore, the aim of this study was to clarify the association between the concentration of ionic components in PM2.5 and respiratory symptoms or PEFR.

2. Materials and Methods

2.1. Study Design

From 1 February to 31 May 2020, we investigated the relationship of PM2.5 and the concentration of its ionic components with PEFR and the daily prevalence of respiratory symptoms in patients with allergic diseases.

2.2. Participants

We started recruiting outpatients with allergic diseases at National Hospital Organization Fukuoka National Hospital (Fukuoka Hospital), in Fukuoka, Japan, in 1 December 2019, and examined patients from February 1 to May 31, 2020. Participants were asked to measure their PEFR with a Mini-Wright PERF meter (Clement Clark International, London, UK) three times at same time every morning and evening and to record the highest measurement in a diary. Before participants started the study, a specialist nurse explained the PEFR measurement method and how to store and clean the PEFR meter and assisted participants with measurements until they could perform them well. Participants were asked also to record their daily respiratory symptoms in their diary. At each outpatient visit, a nurse confirmed that patients were recording symptoms correctly in the diary.

2.3. PM2.5 and Its Ionic Components

PM2.5 and its ionic components were recorded daily by the high-volume air sampler HV-RW-k1 (Shibata Kagaku, Co. Ltd., Tokyo, Japan) on the roof of Fukuoka Hospital.

Every 2 weeks, the filter in the sampler was changed at 10:00 a.m. by a trained inspection engineer and sent to the National Institute for Environmental Studies for measurement of the ionic components. The measurement method of each ion and the details of component analysis were according to the method reported by Yoshino and Takami et al. [8].

Fukuoka City is surrounded by mountains and has a fan-shaped opening towards the sea. Fukuoka Hospital is located at the heart of that fan, so measurements at the hospital are assumed to accurately reflect the air quality in the whole city because the wind from the sea causes the air in the city to gather at the base of the mountains. Fukuoka City has long been a central city in western Japan, with a population of 1.5 million and a commercial city with many company branches. The climate is relatively warm, and such an environment has not changed for a long time. There are no large factories in the city, but cross-border pollution can be seen mainly in early spring.

We chose 1 February through 31 May 2020 as our study period because environmental health surveillance data indicated that the concentration of PM2.5 is relatively high during these months in Japan, especially in this area of the southwest [9]. In China, a five-year plan for air pollution control began in 2013, and air pollution control has progressed in major cities. It was reported that the concentration of fine particulate matter (PM2.5) in Japan decreased as a result. Despite this situation, transboundary pollution is presumed to be the reason for the high concentration of PM2.5 during this period in Fukuoka [10].

2.4. PEFR

We calculated the daily ratio to the morning and evening personal best PEFR (pbPEFR), daily ratio to the morning personal best PEFR (pbPEFRm), and daily ratio to evening personal best PEFR (pbPEFRe) as the ratio of the daily mean morning and evening, morning, or evening PEFR to the personal best mean of morning and evening, morning, or evening PEFR in the study period (1 February to 31 May 2020) with the following formula:

pbPEFR, pbPEFRm, or pbPEFRe = (mean daily, morning, or evening PEFR)/personal best PEFR in the study period

2.5. Respiratory Symptoms

We analyzed the daily prevalence of coughing and nasal symptoms (runny nose, stuffy nose, sneezing) from information recorded by patients in their study diary.

2.6. Statistical Analysis

Statistical analyses were performed with R-3.6.1 (Foundation for Statistical Computing, Vienna, Austria). Univariate linear regression was used to assess the co-relationship between each ionic component and pbPEFR, pbPEFRm, pbPEFRe, and respiratory symptoms with Spearman’s test.

Variables with a skewed distribution were analyzed after logarithmic transformation. To ascertain which variables were independently associated, multiple regression analysis was performed by a retrospective stepwise method in which variables that had a significant relationship with pbPEFR, pbPEFRm, pbPEFRe, or a respiratory symptom in univariate analysis were removed in ascending order of the relationship.

A p value less than 0.05 (two tailed) was considered to indicate a statistically significant difference.

3. Results

From 1 February to 31 May 2020, we recruited a total of 72 patients (male, n = 38; female, n = 34). The mean (SD) age was 16.0 (14.9) years (median, 10 years; range, 2–74 years).

Participants had been diagnosed with the following allergic diseases: asthma (n = 65), atopic dermatitis (n = 29), allergic rhinitis (n = 47), allergic conjunctivitis (n = 4), and sinusitis (n = 7).

3.1. PM2.5 and Its Ionic Components

PM2.5 and concentrations of the ions Cl⁻, NO₃⁻, SO₄²⁻, Na⁺, NH₄⁺, K+, and Mg²⁺ were studied. None of the other ions measured as components of PM2.5 were included in the analyses because they were highly correlated (data not shown). Concentrations of PM2.5 and its ionic components, pbPEFR, pbPEFRm, and pbPEFRe values, and prevalences of respiratory symptoms are shown in

Table 1. Characteristics of concentrations of air pollutants, daily prevalence of respiratory symptoms, and personal best peak expiratory flow rates (pbPEFR).

	Unit	Mean	SD	Median	Min	Max
Air pollutants
PM2.5	μg/m3	14.9220	7.7530	13.5380	2.8630	45.5470
CI−	μg/m3	0.0424	0.0688	0.0149	−0.0004	0.3558
NO3−	μg/m3	1.0020	1.0421	0.6922	0.0092	7.5198
SO42−	μg/m3	2.9016	2.0106	2.2711	0.5938	12.3098
Na+	μg/m3	0.4190	0.1960	0.3893	0.0576	0.9702
NH4+	μg/m3	0.7520	0.8239	0.5192	0.0221	5.8882
K+	μg/m3	0.2733	0.1266	0.2583	0.0845	0.8990
Mg2+	μg/m3	0.3428	0.2776	0.2874	0.0906	2.8840
Symptoms
Runny nose	(%) (1/100)	0.3946	0.1086	0.3784	0.1892	0.6486
Stuffy nose	(%) (1/100)	0.3609	0.1118	0.3514	0.1081	0.6486
Sneezing	(%) (1/100)	0.1641	0.0828	0.1389	0.0000	0.3590
Coughing	(%) (1/100)	0.2859	0.0936	0.2703	0.1282	0.7436
pbPEFR (mean)	ratio	0.8260	0.0131	0.8281	0.7832	0.8490
pbPEFR (morning)	ratio	0.8137	0.0139	0.8136	0.7703	0.8381
pbPEFR (evening)	ratio	0.8377	0.0149	0.8392	0.7954	0.8722

.

3.2. Univariate Analysis

The results of the univariate analysis are shown in

Table 2. Correlation coefficients for the relationship of PM2.5 and ionic concentrations of PM2.5 with daily prevalence of respiratory symptoms and mean personal best peak expiratory flow rates.

Ion	Cough	Runny Nose	Stuffy Nose	Sneeze	pbPEFR	pbPEFRm	pbPEFRe
PM2.5	0.002	0.066	0.085	0.159	0.047	0.099	0.023
CI−	0.226	0.371 **	0.404 **	0.260 **	−0.318 **	−0.169 *	−0.379 **
NO3−	0.076	0.405 **	0.315 **	0.403 **	−0.253 **	−0.075	−0.354 **
SO42−	0.062	0.058	0.065	0.017	0.004	−0.026	0.061
Na+	−0.294 **	−0.355 **	−0.302	−0.256	0.263 **	0.169	0.300 **
K+	−0.087	−0.113	−0.158	0.073	0.237 **	0.189 *	0.257 **
Mg2+	0.016	0.118	0.338 **	0.042	−0.208 *	0.132	−0.222 *

** p < 0.01, * p < 0.05 (Spearman’s test [two tailed]). pbPEFR: Ratio of Peak Expiratory Flow Rate(PEFR) to personal best PEFR in this period (mean), pbPEFRm: Ratio of PEFR to personal best PEFR(morning), pbPEFRe: Ratio of PEFR to personal best PEFR(evening).

. The table shows the correlation coefficients for the relationship of PM2.5 and its ionic components with pbPEFR, pbPEFRm, pbPEFRe and the daily prevalence of each respiratory symptom.

An interesting point is that in this study, some of the ions in PM2.5 worsened symptoms and PEFR, while others improved them.

Figure 1 shows a scatter plot of the relationship between NO₃⁻ and pbPEFR, pbPEFm, pbPEFe, and the daily prevalence of respiratory symptoms as a typical example of the relationships.

Vertical axes show the number of symptoms per day in the total number of participants, and horizontal axes show the daily ion concentration of NO₃⁻ in μgm⁻³.

Numbers are logarithmic.

pbPEFR, mean the ratio of daily Peak Expiratory Flow Rate (PEFR) to personal best PEFR in the study period, pbPEFRm, mean morning PEFR, pbPEFRe mean evening PEFR, pbPERF mean the daily average PEFR.

3.3. Multivariate Analysis

We performed a multivariate analysis of the items that were significant in the univariate analysis. For the respiratory symptoms, R² was not much higher than 0.2, and we did not obtain a prediction formula for respiratory symptoms. However, we did obtain a formula for pbPEFR, pbPEFRm, and pbPEFRe (

Table 3. Results of multivariate analysis of mean evening personal best peak expiratory flow rate.

(1) pbPEFR (mean)
Residuals:	Min	1Q	Median	3Q	Max
	−0.034975	−0.007397	0.001305	0.007532	0.026191
Coefficients:	Estimate	Std.Error	t value	P (>1t1)
(Intercept)	0.820187	0.003365	243.755	0.000000000000002 ***
NO3−	−0.003430	0.001267	−2.708	0.007784 **
Na+	0.021338	0.006101	3.497	0.000667 ***
K+	0.025214	0.009322	2.705	0.007866 **
Mg2+	−0.552945	0.214320	−2.580	0.011128
Residual standard error: 0.01142 on 116 degrees of freedom Multiple R-squared: 0.2609, Adjusted R-squared: 0.2355 F-statistic: 10.24 on 4 and 116 DF, p-value: 0.0000003899
(2) pbPEFR (morning)
Residuals:	Min	1Q	Median	3Q	Max
	−0.038602	−0.007319	0.001257	0.008538	0.029258
Coefficients:	Estimate	Std.Error	t value	P (>1t1)
(Intercept)	0.8103733	0.0033654	240.797	0.00000000000000002 ***
PM	0.0003887	0.0001919	2.025	0.045102 *
Na+	0.0214533	0.0066828	3.210	0.001712 **
Mg2+	−0.9597205	0.2557364	−3.753	0.000274 ***
Residual standard error: 0.01307 on 117 degrees of freedom Multiple R-squared: 0.1342, Adjusted R-squared: 0.112 F-statistic: 6.043 on 3 and 117 DF, p-value: 0.0007326
(3) pbPEFR (evening)
Residuals:	Min	1Q	Median	3Q	Max
	−0.0329	−0.0076	0.00129	0.00762	0.0329
Coefficients:	Estimate	Std.Error	t value	P (>1t1)
(Intercept)	0.08303	0.00369	225.134	0.00000000000000002 ***
NO3−	−0.0054	0.0014	−3.876	0.000176 ***
Na+	0.0242	0.0067	3.619	0.00044 ***
K+	0.0336	0.0102	3.286	0.00135 **
Ca2+	−0.5473	0.2349	−2.330	0.02154 *
Residual Standard error: 0.0125 on 116 degrees of freedom Multiple R-squared: 0.3208, Adjusted R-squared: 0.2973 F-statistic: 13.7 on 4 and 116 DF, p value: 0.0000000035

*** p < 0.001, ** p < 0.01, * p < 0.05.

); the results of pbPEFRe are shown in This Table and the respective prediction formula is shown below;

pbPEFRe = 0.8303 − 0.0054NO₃⁻ + 0.0242Na⁺ + 0.0336K⁺ − 0.5473Mg²⁺

Figure 2 shows the relationship between the values calculated by this prediction formula and the measured values and indicates that the formula showed a good correlation (n = 121; rs = 0.626, p =0.000000000000017 [two tailed]).

4. Discussion

This study found that on days with higher concentrations of PM2.5 and its ionic components, a higher proportion of patients with allergic diseases show respiratory symptoms and have a decrease in pbPEFR. The results of multivariate analysis also showed that the concentrations of some ions was associated with a decrease in pbPEFR.

The measured value of pbPEFRe showed a good correlation with the value calculated by the prediction formula obtained from multivariate analysis, which included the ion concentrations on that day. In this study, we used pbPEFR, a PEFR metric that is slightly different from the commonly used %PEFR and has not been used much in the past. However, the metric may be useful for studying changes in PEFR in people with different background factors, such as those with respiratory diseases and children. We have previously published findings of studies that used this index and showed its usefulness [11,12].

Recently, an increasing number of studies have evaluated the effects of PM2.5 components on health [13,14].

When considering the influence of PM2.5, it seemed necessary to consider the increase and decrease of each component and its influence. The effects of this study were also mixed.

Therefore, we considered it necessary to perform an additional study on the health effects of PM2.5 and its ionic components.

In 2020, there was the coronavirus pandemic. In Japan, the first case was found on 28 January. After that, on 28 February, the Japanese government declared a state of emergency in seven prefectures, including Fukuoka, which was extended until the end of May. According to a survey by the Japanese Society of Pediatrics, the number of outpatients began to decrease in March, reaching its lowest point in May, and then began to increase. In October, the number of patients returned to pre-COVID-19 levels, but acute diseases such as bronchitis are said to have remained low [15].

In western Japan, including Fukuoka, the amount of PM2.5, including yellow dust, from the continent is high from March to May due to air currents. In 2020, it is still an issue to be examined whether the corona epidemic affected its composition and concentration. In addition, it is presumed that the visits of patients at hospitals and clinics were affected. The period we investigated overlapped with a fluctuating and complicated period in which the number of patients visiting hospitals decreased, the economy deteriorated, and some industries showed signs of recovery.

However, it is interesting that even in this fluid period, a statistically significant association was observed, and that associations common to previous reports such as NO3- were observed. Waiting for further investigation is likely to be a necessary but meaningful result.

It is important to consider whether the PM2.5 values obtained in this study can be considered typical for this region in a typical corona-free year, or as a special case only for 2020. This time, the joint researchers are conducting research including the flow of air currents from the continent. However, unfortunately, it is difficult to make such a judgment at this time, taking into account many factors.

One of the limitations of this study is the small number of cases, which was due in part to the coronavirus pandemic. After that, it started to recover from June. According to the report [13], from March to May, the number of hospital visits for acute illness decreased by about half, and that for chronic illness decreased by 70%. In the future, it may be necessary to confirm the results in a larger study.

As mentioned in the section on object and method, Fukuoka is an environmentally stable city without major disasters and is a stable city when considering the environment as a background factor. Of course, there are advantages to conducting surveys in multiple regions, but it is thought that there are advantages to conducting surveys under stable background factors in a certain region [16,17,18,19,20].

There are still few reports on the relationship between the ionic components of PM2.5 and the pathology of asthma. A report examined the short-term association between particulate matter (PM2.5) concentration and its components, and hospital emergency room visits (ERV) for asthma in southern Taiwan. In this study, asthma ERV were reported to be significantly associated with nitrate (NO₃⁻) concentrations [21]. There are also reports that the influence of sulfate in the atmosphere is not so strong [7], and this is also the case with our results. These are interesting because it agrees with our report.

On the other hand, the importance of the influence of the carbon component of PM2.5 has been pointed out [7], but we could not examine it this time. In this study, a certain degree of prediction formula was obtained for pbPEFRe, but it may be possible to make better predictions by adding factors to the study in the future. We would like to consider this in the future.

The strength of this study is that it examined the effects of PM2.5 components on the rate of changes in symptoms and lung function in the same patient population over a 4-month follow-up period, and such reports are rare. To our knowledge, this was the first study on ions in particulate matter in this methods, and we believe that some of the results will be useful for future research. Further studies need to be conducted in multiple regions and to examine participant background factors. Whether similar results are obtained by studies in other regions remains to be seen, but such research often requires an epidemiological approach. Once a suitable epidemiological method has been established for a study in one area, the next step would be to repeat the study in other areas and verify the approach. We hope that such studies are performed in the future.

The strength of this study is that it investigated the effects of the ionic components of PM2.5 on lung function and respiratory symptoms, an area that has been rarely studied. We hope that our findings will lead to increased research on this topic.

The fact that the lung function and respiratory symptoms of people with allergic diseases are affected by the ionic components in PM2.5 provides extremely useful information for the management and treatment of clinical symptoms in these patients.

An interesting point to note is that in this study, some of the ions in PM2.5 worsened symptoms and PEFR, while others improved them. These findings need further confirmation in future studies, which should also examine the health effects, including the associated mechanisms.

5. Conclusions

Among the ionic components of PM2.5, Cl⁻, NO₃⁻, Na⁺, K⁺, and Mg²⁺ are significantly correlated with the frequency of respiratory symptoms such as coughing, sneezing, runny nose, and stuffy nose and pbPEFR, pbPEFRm, and pbPEFRe in univariate analysis. As shown by multivariate analysis, changes in individual pbPEFRe can be predicted to a certain extent from the concentration of ions, NO₃⁻, Na⁺, K⁺, and Mg²⁺.

We concluded that some ionic components of PM2.5 may increase the frequency of respiratory symptoms and decrease PEFR in patients with allergic diseases, and may predict PEFR to some extent.