Statistical Analysis
Analyses were performed using SPSS Statistics v21.0 (IBM, Chicago, IL, USA). Normally-distributed continuous data were expressed as mean and standard deviation, and not-normally distributed continuous data as median and interquartile ranges (IQR). Group comparisons were carried out using student´s t-test, Mann–Whitney U-test or ANOVA as appropriate for the continuous, and the chi-square test or Fisher test for categorical variables. The correlation coefficients between serum periostin level and other clinical variables were determined using Spearman‘s rank correlation coefficient. Association of severe asthma classes with clinical variables was examined using regression models adjusted for potential confounders, including age, aeroallergen sensitization, BMI z-score, blood eosinophil count and serum periostin level. The odds ratio (OR) and 95% confidence interval (CI) were reported. The diagnostic performances of serum periostin levels to identify children with severe asthma were determined by receiver operating characteristic (ROC) curve analysis. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for the selected cut-off point. A P level <0.05 was considered significant.