Atopic dermatitis severity during exposure to air pollutants and weather changes with an Artificial Neural Network (ANN) analysis.

Abstract

Epidemiological studies have shown an association between global warming, air pollution, and allergic diseases. Several air pollutants, including volatile organic compounds, formaldehyde, toluene, nitrogen dioxide (NO2 ), and particulate matter, act as risk factors for the development or aggravation of atopic dermatitis (AD). We evaluated the impact of air pollutants and weather changes on AD patients. Sixty AD patients ≥5years of age (mean age: 23.5±12.5years), living in the Campania Region (Southern Italy), were followed for 18months. The primary outcome was the effect of atmospheric and climatic factors on signs and symptoms of AD, assessed using the SCORAD (SCORing Atopic Dermatitis) index. We measured mean daily temperature (TOD), outdoor relative humidity (RH), diurnal temperature range (DTR), precipitation, particulate with aerodynamic diameter≤10μm (PM10 ), NO2 , tropospheric ozone (O3 ), and total pollen count (TPC). A multivariate logistic regression analysis was used to examine the associations of AD signs and symptoms with these factors. An artificial neural network (ANN) analysis investigated the relationships between weather changes, environmental pollutants, and AD severity. The severity of AD symptoms was positively correlated with outdoor temperatures (TOD, DTR), RH, precipitation, PM10 , NO2 , O3 , and TPC. The ANN analysis also showed a good discrimination performance (75.46%) in predicting disease severity based on environmental pollution data, but weather-related factors were less predictive. The results of the present study provide evidence that weather changes and air pollutions have a significant impact on skin reactivity and symptoms in AD patients, increasing the severity of the dermatitis. The knowledge of the single variables proportion on AD severity symptoms is important to propose alerts for exacerbations in patients with AD of each age. This finding represents a good starting point for further future research in an area of increasingly growing interest.