Comment on “Distribution features of biological hazardous pollutants in residential environments in Korea [Lee et al., Environ Sci Pollut Res (2014) 21:1146–1152]”

In their article, Lee et al. (2014) report levels of the major group 1 allergens from the house dust mites Dermatophagoides pteronyssinus and Dermatophagoides farinae, and bacterial endotoxin in bedding from pregnant women and 6-month-old newborns. There are a number of methodological and interpretive issues with their study. First, endotoxin was determined by the Limulus amebocyte lysate (LAL) method in aqueous extracts of bedding dust. Most, if not all, researchers extract dust samples or filters with pyrogen-free water with the addition of 0.05 % Tween-20 as Douwes et al. (1995) has shown that endotoxin extraction efficiency is about seven times higher in Tween-20containing water than in pyrogen-free water only. Thus, the endotoxin results reported by Lee et al. (2014) maybe seriously underestimated and this may be why the authors only found a detection ratio of endotoxin of 94.5 %, unlike the detection ratio of 100 % in other studies. Secondly, house dust allergen and endotoxin data is skewed, and most studies have log transformed this data before statistical analysis and report results as geometric mean levels with geometric standard deviations or 95 % confidence intervals. It therefore makes it difficult to compare the data of Lee et al. with other published studies. Also, due to likely skewed distribution of allergen and endotoxin data, statistical analysis should have been done by appropriate non-parametric tests. Lee et al. (2014) sampled dust from mattresses in homes with beds, but in homes without beds, dust was collected from blankets and bedclothes. The authors did not present data on the number of samples from the items other than mattresses. This makes interpretation of their data difficult or possibly misleading as it is well known that there are major differences in levels of allergens and endotoxin between mattresses and other bedding items (Hall et al. 2002; Mills et al. 2002; Wickens et al. 2004), with mattresses generally having much higher levels. Lee et al. (2014) found detection rates ofDer p 1 and Der f 1 of 52.7 and 86.5 %, respectively, and deduce thatD. farinae is the dominant house dust mite species in their locality. We have previously shown that D. farinae is also the dominant house dust mite species in Cheonan, Korea (Nam et al. 2008). Our detection rates differed from Lee et al. (2014); in that, we were able to detect Der f 1 in all (100 %) of mattresses, but Der p 1 in only 12 % of mattresses. These differences in detection rates between the two studies in Korea may be due to the nature of the sampled items. In our study (Nam et al. 2008), we only sampled mattresses, while Lee et al. (2014) studied a mix of mattresses and other bedding items.