Ionic Components of Particulate Matter 2.5 May Influence Daily Prevalence of Skin Symptom Exacerbations in Allergy Sufferers

Featured Application

The present study found a significant relationship between the prevalence of skin symptom exacerbations and the concentration of particulate matter 2.5 and its ionic components. A few studies have evaluated this topic, so the present study may be an important stepping stone for future research.

Abstract

(1) Background: To date, little research has epidemiologically examined whether the concentration of particulate matter (PM) 2.5 and its ionic components is related to the prevalence of skin symptom exacerbations. Therefore, this study aimed to clarify this association in patients with allergic diseases. (2) Methods: From 1 February to 31 May 2020, we evaluated the daily prevalence of skin symptoms in outpatients with allergic diseases being treated at Fukuoka National Hospital, Fukuoka, Japan, and measured the concentration of PM2.5 and its ionic components. (3) Results: Univariate analysis showed a statistically significant association between skin symptoms and the concentration of PM2.5 and the ionic components SO₄²⁻, NH₄⁺, K⁺, and Mg²⁺; multivariate analysis showed a statistically significant association between the daily prevalence of skin symptom and the concentration of the ionic components SO₄²⁻ or Mg²⁺. (4) Conclusions: Our findings indicate that the concentration of some PM2.5 ionic components may affect skin symptom in patients with allergic diseases.

1. Introduction

Air pollutants have been reported to affect allergic diseases, especially airway diseases [1]. Previously, we reported a relationship between airway symptoms and PM2.5 and its ionic components [2]. In clinical practice, patients with allergic diseases sometimes say that their skin symptoms are worse on days with a high concentration of air pollutants. Research reports on air pollution and skin have gradually increased in recent years [3]. However, there are currently few reports on the relationship between PM2.5 components, especially ionic components. By researching ionic components and other components of PM2.5, it may be possible to consider the causes and effects of PM2.5, even if it is the same PM2.5, as well as regional differences, and it may be possible to consider the health effects in more depth. However, few studies have evaluated the effects of air pollution [4] on the prevalence of skin symptoms [3]. Furthermore, to the best of our knowledge, few studies have examined whether the specific ionic components of PM2.5 affect skin symptoms.

Because PM2.5 and its ionic components are associated with the prevalence of respiratory symptoms, we hypothesized that they may also be associated with the prevalence of skin symptom exacerbations in patients with allergic diseases. Therefore, this study aimed to clarify the association between the prevalence of skin symptom exacerbations and the concentration of PM2.5 and its ionic components in such patients.

2. Materials and Methods

2.1. Study Design

From 1 February to 31 May 2020, we performed a prospective cohort study to investigate the relationship between the daily prevalence of skin symptoms in patients with allergic diseases and the concentration of PM2.5 and its ionic components. This study was approved by the ethics committee of the National Hospital Organization Fukuoka National Hospital (Fukuoka Hospital) (No. F31-03, date of approval: 15 May 2019) and was performed in accordance with the Declaration of Helsinki.

2.2. Participants and Study Procedures

We started recruiting outpatients with allergic diseases being treated at Fukuoka Hospital in Fukuoka, Japan, on 1 December 2019, and examined patients from 1 February to 31 May 2020. Participants gave written informed consent to participate in this study.

Participants were given a diary and asked to use it to record their daily skin symptoms and to answer the question “Did your skin get worse?” At each outpatient visit, a specialist nurse confirmed that participants were recording their symptoms correctly.

At the end of the data collection period, the proportion of participants who recorded a worsening of symptoms (defined as answering “yes” to the question of whether their skin had worsened) was analyzed as the daily prevalence of skin symptoms.id

2.3. Measurement of PM2.5 and Its Ionic Components

PM2.5 and its ionic components were measured daily using a large-capacity HV-RW-k1 air sampler (Shibata Scientific Co., Ltd., Tokyo, Japan) installed on the rooftop of Fukuoka Hospital. Sampler filters were changed daily at 10 a.m. by trained laboratory technicians and stored at −20 degrees Celsius. The filters were sent to the National Institute for Environmental Studies every two weeks to measure PM2.5 and its ionic components. Details of the PM2.5 measurement method and component analysis are provided in a separate report [5].

The study period was set from 1 February to 31 May because, according to environmental health monitoring data, PM2.5 concentrations in this region were particularly high during this period [6]

Measuring pollutant concentrations for use in investigating the health effects of air pollutants ideally requires each individual subject to carry a measuring device and be monitored. However, carrying out such monitoring in actual clinical cases poses many difficulties; therefore, pollutant concentration measurements in this study were carried out on the roof of Fukuoka Hospital. Measurements were carried out daily by experienced laboratory technicians. The idea was to verify predictions to a certain extent using this method and to conduct more detailed investigations if necessary. At this stage, a certain tendency was identified between the ionic components of pollutants and skin symptoms, so the results are reported here.

2.4. Statistical Analysis

Statistical analyses were performed using R-3.6.1 (R Foundation for Statistical Computing, Vienna, Austria). First, we performed univariate linear regression using Spearman’s test to examine the relationship between PM2.5 or its ionic components and skin symptom exacerbations. Next, in order to ascertain which variables were independently associated with skin symptoms, a retrospective stepwise analysis was performed in which variables found to be significantly associated with skin symptoms in univariate analysis were excluded in order of their relevance. A multiple regression method was used.

Differences were considered statistically significant if a p-value (two-tailed) less than 0.05 was obtained.

3. Results

From 1 December 2019 to 31 January 2020, we screened 78 patients, 72 of whom (male, n = 38; female, n = 34) were eligible for this study and agreed to participate from 1 February to 31 May 2020. Of the 78 patients, 6 were excluded because they had no allergic disease.

The characteristics of the subjects are shown in

Table 1. Characteristics of the subjects.

				n			Average		SD		Median			Min		Max
Age(years)		Male		38			12.8		10.3		10.0			4.0		50.0
		Female		34			19.2		16.3		11.0			3.0		58.0
BL (cm)		Male		38			145.6		19.9		150.5			99.0		173.0
		Female		34			140.3		21.1		145.5			87.3		168.0
BW(kg)		Male		38			42.5		17.3		37.5			15.2		80.0
		Female		34			37.4		14.9		39.0			13.2		74.0
Allergic Diseases
			BA		AD	AR		FA		AC		U	P		Others		Total
Male	(n)		37		13	19		5		6		2	3		12		38
	(%)		97.4		34.2	50.0		13.2		15.8		5.3	7.9		31.6
Female	(n)		30		10	18		4		6		7	1		8		34
	(%)		88.2		29.4	52.9		11.8		17.6		20.6	2.9		23.5

SD: standard deviation, Min: minimum, Max: maximum, BL: body length, BW: body weight, BA: bronchial asthma, AD: atopic dermatitis, AR: allergic rhinitis, FA: food allergy, AC: allergic conjunctivitis, U: urticaria, P: pollinosis.

.

3.1. PM2.5 and Its Ionic Components

The characteristics of PM2.5 and its ionic components are shown in Table 1.

3.2. Univariate Analysis

The results of the univariate analysis of the relationship between the prevalence of skin symptoms and the concentrations of PM2.5 and its ionic components(

Table 2. Concentrations of PM 2.5 and its ionic components measured from January 1 to April 31, 2020, at Fukuoka National Hospital, Fukuoka, Japan.

	Unit	Mean	SD	Median	Min	Max
PM2.5	μg/m3	14.9	7.75	13.5	2.86	45.55
Cl−	μg/m3	0.04	0.07	0.015	0.00	0.356
NO3−	μg/m3	1.00	1.04	0.69	0.01	7.52
SO42−	μg/m3	2.90	2.01	2.27	0.59	12.31
Na+	μg/m3	0.42	0.20	0.39	0.06	0.97
NH4+	μg/m3	0.75	0.82	0.52	0.02	5.89
K+	μg/m3	0.27	0.13	0.26	0.08	0.90
Ca2+	μg/m3	0.01	0.01	0.01	0.004	0.059
Mg2+	μg/m3	0.34	0.28	0.29	0.09	2.88

SD: standard deviation, Min: minimum, Max: maximum.

) are shown in

Table 3. Relationship between the daily prevalence of skin symptom exacerbations and the concentration of PM2.5 and its ionic components.

	rs	P
PM2.5	0.3905 **	9.50 × 10−6
Cl−	0.0343	0.7084
NO3−	0.1014	0.2684
SO42−	0.3263 **	0.0003
Na+	0.1525	0.095
NH4+	0.3214 **	0.0003
K+	0.2292 *	0.0115
Ca2+	0.0651	0.4778
Mg2+	0.3018 **	0.0008

PM2.5, particulate matter 2.5; rs, Spearman’s rank correlation coefficient; * p < 0.05, ** p < 0.01, Spearman’s test, n = 121, two-tailed.

and Figure 1. We found a statistically significant relationship between the prevalence of daily skin symptoms and the concentrations of PM2.5, SO₄²⁻, NH₄⁺, K⁺, and Mg²⁺.

3.3. Multivariate Analysis

We performed a multivariate analysis of the items that were significant in the univariate analysis (Table 3). Neither PM2.5 nor any of the ionic components other than SO₄²⁻ and Mg²⁺ were included in the analysis because they were highly correlated with PM2.5 or these other ions (as shown in

Table 4. Correlation coefficients between each variable.

	PM2.5	CL−	NO3−	SO42-	Na+	NH4+	K+	Ca2+	Mg2+
PM2.5	-
Cl−	0.209	-
NO3−	0.533	0.148	-
SO42-	0.554	−0.2	0.387	-
Na+	0.212	0.129	0.071	0.416	-
NH4+	0.598	−0.157	0.665	0.919	0.22	-
K+	0.601	−0.285	0.384	0.721	0.188	0.753	-
Ca2+	0.594	0.592	0.458	0.025	0.416	0.231	0.065	-
Mg2+	0.647	0.422	0.393	0.215	0.297	0.179	0.143	0.815	-

, there is a strong correlation between Mg²⁺ and Ca²⁺ ions, and between NH₄⁻ and SO₄²⁻ ions, so Ca²⁺ and NH₄⁻ were excluded from the calculation).

According to the results of the multivariate analysis, the prediction formula for the prevalence of skin symptoms was as follows: prevalence of skin symptom exacerbations = 0.538 + 0.0017 × SO₄²⁻ + 0.00108 × Mg²⁺.

The relationship between the values calculated by this formula and the measured values was statistically significant (n = 121, p = 0.00026, r = 0.326; Figure 2,

Table 5. Results of multivariate analysis of the daily prevalence of skin symptoms and the concentrations of PM2.5 and ionic components that showed a significant relationship in univariate analysis.

Residuals	Min	1Q	Median	3Q	Max
	−0.018841	−0.006410	−0.000877	0.005179	0.030017
x	Estimate	Std. error	t-value		Pr(>ltl)
(Intercept)	0.5380167	0.0018365	292.956		<2 × 10−16 ***
SO42−	0.0016986	0.0004693	3.620		0.000436 ***
Mg2+	0.0108384	0.0033995	3.188		0.001833 **

1Q, first quartile; 3Q, third quartile. Residual standard error: 0.01009 on 118 degrees of freedom. Multiple R-squared, 0.2005; adjusted R-squared, 0.1869. F-statistic, 14.79 on 2 and 118 DF; p-value = 1.848 × 10⁻⁶. ***: p < 0.001, **: p < 0.01

).

The prevalence of skin symptoms was calculated by the following formula obtained from multivariate analysis: 0.538 + 0.0017 × SO₄²⁻ + 0.0108 × Mg²⁺.

The relationship between the skin symptoms calculated by this formula and the daily skin symptoms recorded by patients was statistically significant (n = 121, rs = 0.326, p = 0.00026 [two-tailed]). These results indicate that daily skin symptoms depend to some extent on the concentrations of SO₄²⁻ and Mg²⁺ in PM2.5.

4. Discussion

This study found an association between the prevalence of everyday skin symptoms in patients with allergic diseases and the overall concentration of PM2.5 and some of its ionic components.

4.1. Air Pollution and Allergic Diseases

Allergic diseases have been on the rise over the past several decades, but asthma and atopic dermatitis have not increased in Japan over the past 10 to 20 years [7]. Although the reasons for the increase in allergic diseases are not clearly understood, we reported that the prevalence of childhood asthma may be related to the concentration of suspended particulate matter in the atmosphere, including PM2.5 [8]. We also reported that the ionic component concentration of PM2.5 was associated with nasal symptoms, respiratory symptoms, and maximum expiratory flow rate [2]. We sometimes hear from patients with atopic dermatitis that their skin symptoms worsen on days when PM2.5 concentrations are high. However, no epidemiological investigation was conducted, and this issue was ignored. Based on the above, we hypothesized that air pollution may be related to the symptoms of atopic dermatitis and that daily changes in skin symptoms may be related to changes in the concentration of PM2.5 and its ionic components.

4.2. Relationship between Air Pollutants and Skin Symptoms

There are not many studies that have evaluated the relationship between skin symptoms and air pollutants, especially PM2.5, which has recently become a problem, and its ionic components, or the impact of these air pollutants on allergic diseases other than asthma. Several studies have been conducted on the effects of air pollutants on dermatitis. It has been reported that the prevalence of eczema increases when NO₂ concentrations are high and patients live near major roads (e.g., highways) [9,10,11]. A Korean study reported that exposure to PM2.5 increases eczema symptoms [12]. It has been reported that the prevalence of allergic diseases decreases by 0.39% when the NO₂ concentration decreases by 1 ppb and by 0.14% when the suspended particulate matter concentration decreases by 1 mg/m³ [13].

There are many reports on the relationship between air pollution and allergic diseases. However, many of these reports show comparisons of disease prevalence in areas or days with high or low concentrations of air pollutants.

This report investigated daily changes in disease prevalence and air pollution concentrations. There are not many reports from this perspective. Research from this perspective requires a high degree of cooperation from participants, but since the data come from the same person, it is unlikely that the background will change. This may be the strength of this study.

4.3. Impact of the COVID-19 Pandemic

This study was conducted from February to April 2020. Markham et al. [14] compared pediatric emergency department visits from March to August 2020 during the COVID-19 pandemic with visits in the same months in the three years before the pandemic and found that hospitalizations for organic diseases significantly decreased. It has been reported that the number of common organic diseases decreased significantly. The results of that epidemiological study suggest that the prevalence of respiratory diseases may have been influenced by conditions associated with the pandemic, such as reduced traffic. This suggests that protective measures to prevent the spread of COVID-19, such as social distancing and wearing masks, may have been effective in reducing the spread of the virus and bacteria.

Previously, we also reported that mask wearing had a positive effect on symptoms in patients with allergic diseases [15]. Wearing a mask was effective in reducing the prevalence of respiratory symptoms and improving peak expiratory flow. However, even under such circumstances, the effects of PM2.5 and its ionic components on skin symptoms have been confirmed.

The COVID-19 pandemic has also been impacting allergies through changes (or restrictions) to daily life, such as reduced attendance at school or work, fewer opportunities to meet others, and less travel. However, changes in daily life during this study period were unlikely to have affected the effects of air pollutants on the skin.

4.4. Limitations

One of the limitations of this study is that participants reported their own symptoms, so the data are subject to self-report bias.

One further limitation of this study is the small number of participants. However, our study has the advantage of being conducted over a certain period of time on the same patients, and statistically significant results were obtained. In fact, we hope that the weaknesses of this study will be overcome in the next study. It was also conducted in a limited area. However, a positive correlation was observed between skin symptoms and the concentrations of some air pollutants. Although studies in a limited area may have problems with universality, they are also considered to be advantageous for seeing the effects over a certain period of time. However, future studies in many areas are needed.

A limitation of this study may be the lack of detailed and strict diagnostic criteria for subtle changes in skin lesions. However, this time, we investigated questions regarding the worsening of skin symptoms to verify the hypothesis that PM2.5 and its ionic components may affect the overall skin symptoms in patients with allergic diseases. This study revealed that PM2.5 and its ionic components are likely to affect skin symptoms. Furthermore, it was found that this effect has at least a short-term effect (acute effect).

4.5. What Can We Learn from This Study?

We have already reported that the prevalence of AD was on the decline in 2002 [16], and a survey of first-grade elementary school students in Himeji City, Japan also showed a clear decrease [17]. Yura et al. found that, according to a large-scale survey of first-year elementary school students, the increase in AD had stopped since the 1990s, and there were no longer any differences between grades [18]. As mentioned above, epidemiological surveys in Japan have shown that the prevalence of AD is on the decline. This trend may be explained by the decreasing trend in air pollution, especially PM2.5 and its ion concentrations, in Japan. However, we believe that the prevalence of AD is not solely determined by air pollution. The results of this study show that the concentrations of PM2.5 and some of its ionic components have a statistically significant positive correlation with the prevalence of AD. This suggests that PM2.5 and its ionic components may be important factors in AD. As AD worsens, the incidence of symptoms increases. Furthermore, regarding the daily prevalence of AD worsening, a significant correlation was observed between the measured value and the calculated value using the formula obtained from the multivariate test. This indicates that the symptoms of AD worsen as the concentration of PM2.5 and its ionic components increases, and repeated symptoms are known to be an exacerbating factor. This is presumed to be related to the prevalence of the condition. It is possible that the recent decline in the prevalence of AD may be related to the decline in PM2.5 concentration.

Our findings warrant future studies using more stringent evaluation criteria and including a more detailed analysis of the relationship between the proportion of all ionic components of PM2.5 and skin symptoms. Furthermore, if future research reveals the effects of ionic components on the skin, it will become clear whether the effects are simply physical stimulation or chemical effects, which may lead to the prevention of skin symptoms. It is believed that great effects can be expected not only in understanding but also in improving the treatment.

5. Conclusions

We investigated the effect of PM2.5 and its ionic components on the prevalence of daily skin symptom exacerbations in outpatients with allergic diseases. Univariate analysis showed a statistically significant short-term association between skin symptoms and the concentrations of PM2.5 and its ionic components. Multivariate analysis showed a significant association between skin symptoms and SO₄²⁻ and Mg²⁺, suggesting that it is important to examine the relationship between skin symptoms and these ionic components.