The effect of sieving on Der p 1 levels.

Abstract

. THE house-dust mite Dermatophagoides pteronyssinus produces a major group 1 allergen (Der p 1) which is associated with the development and exacerbation of atopic asthma (1). Collection of dust samples by vacuuming floors and bedding is an accepted technique for assessment of domestic exposure to Der p 1. As well as recommendations for standardized dust collection techniques, it has also been recommended that large particles be removed by sieving before extraction and analysis for Der p 1 (1). This can be a problem when dust samples are low in weight, such as those collected from smooth flooring or individual bedding items, such as pillows. The aim of this study was to investigate whether Der p 1 concentration is affected by sieving dust before analysis. Dust was collected from 24 carpeted living-room floors and from 24 mattresses by vacuuming 1 m for 1 min with a Hitachi CV-2500 vacuum cleaner (1100 W). After thoroughly mixing the sample, an aliquot (0.09–0.11 g) was removed, weighed, and coded. The remaining dust sample was sieved through a 425-mm steel mesh sieve, and an aliquot (0.09–0.11 g) of the fine sifted dust was removed, weighed, and coded. The unsieved and sieved dust samples were extracted with 1 ml of phosphate-buffered saline (containing 0.05% Tween-20) for 30 min at room temperature (2), and Der p 1 was measured in the supernatant by established double-monoclonal antibody ELISA (3). Der p 1 levels were logtransformed and expressed as geometric means with 95% confidence intervals (95% CI). Comparison of Der p 1 results was by linear regression analysis, and the differences between unsieved and sieved Der p 1 results were plotted against their mean values, as outlined by Bland & Altman (4). Geometric mean Der p 1 (95% CI) for floor dust samples were 17.2 mg/g (9.8–30.2) and 19.5 mg/g (11.3–35.6) for unsieved and sieved dust samples, respectively. Corresponding levels for mattress dust samples were 57.9 mg/g (38.6–86.9) and 66.0 mg/g (43.4–100.3) for unsieved and sieved dust samples, respectively. These results were not statistically significantly different. Der p 1 levels were, on average, 16.4% (95% CI: 4.128.8) and 15.5% (95% CI: 6.9-24.2) higher in sieved dust samples from floors and mattresses, respectively, compared to corresponding unsieved dust samples. Comparison of Der p 1 concentrations of unsieved and sieved dust samples yielded Pearson correlation coefficients (r) of 0.966 (y=0.951x+0.114) and 0.986 (y=1.017x+0.027), respectively, for floor and mattress dust samples. For floor dust samples, the mean difference was –4.18 mg/g, with limits of agreement of –20.44 to 12.08 mg/g. Correspondingly, for mattress dust samples, the mean difference was –10.59 mg/g, with limits of agreement of –36.67 to 15.49 mg/g. For both floor and mattress dust samples, the differences exceeded the 95% confidence limits in only one of the 24 dust samples. This study has shown that, although yielding lower Der p 1 levels, not sieving dust did not result in significantly different Der p 1 concentrations, compared to sieving dust. Although only dust samples from floors and mattresses were compared in this study, we believe the results are applicable to other commonly sampled items, such as clothing and pillows, where the amount of dust collected is often small (5). Sieving dust takes time, and is an additional factor that may contribute to the total inaccuracy inherent in measuring Der p 1 levels in reservoir dust.