Association between asthma severity and periostin
There are also inconsistent results from the studies which investigated the association between asthma severity and serum periostin levels. The results of Licari et al.(17) and Konradsen et al. (18) did not reveal an association between asthma severity and serum periostin levels. On the other hand, Song et al. (9) have explored the relationship between periostin and airway hyperresponsiveness (AHR) in children with asthma and found a significant correlation between the degree of AHR and periostin levels. Similarly, Cho et al. (19) found significantly higher periostin levels in children with positive exercise and mannitol tests when compared to children with asthma and negative results. Recently, the findings of Knihtilä et al.(20) have demonstrated a significant correlation between serum periostin levels, airway-hyperreactivity and bronchodilator responsiveness in 49 children with asthmatic symptoms. In our study we found a significant and independent correlation between asthma severity and serum periostin levels. A serum periostin value of 52 ng/ml was emerged as the best cut-off level to differentiate children with severe asthma with high sensitivity and negative predictive values, whereas the specificity and positive predictive value for this cut-off were not satisfactory. According to our findings, it can be postulated that low serum periostin levels may serve clinicians better in excluding severe asthma in children.