Developing a faster way to identify biocontamination in the air of controlled environment rooms with HEPA filters: airborne particle counting

Abstract

Our aim was to assess whether airborne particle counting is an immediate indicator of biocontamination in controlled environment rooms with HEPA filters in a hospital. A prospective study was carried out in a tertiary care hospital between 2016 and 2018. The study was divided in two periods and the measurements were performed in different controlled environment rooms with HEPA filters. The Environmental Biosafety Criterion (EBC) was defined as the absence of fungal and bacterial contamination. In the training period, the area under the ROC curve (aROC) of airborne particle counting and EBC was calculated for each particle size as well as the cut-off points that optimize the combination of sensitivity and specificity in the association between them. aROC is created by plotting sensitivity against 1-specificity. In the testing period, the cut-off points previously selected were validated. 328 measurements were carried out in the training period and 301 in the testing period. In the training period, an association was found between airborne particle counting and EBC. An aROC = 0.760, 95% Confidence Interval (95% CI) 0.695-0.825 was observed for 0.3 µm particles; an aROC = 0.797 (95% CI 0.734-0.860) for 0.5 µm particles; and an aROC = 0.751 (95% CI 0.673-0.829) for 5 µm particles. The cut-off points that optimized the combination of sensitivity and specificity were 9.0 × 10³ for 0.3 µm particles, 3.6 × 10³ particles for 0.5 µm, and 3.2 × 10² particles for 5 µm. In the testing period, the previous cut-off points were validated. We conclude that airborne particle counting is a useful, immediate, and preliminary measure to identify the presence of biocontamination in controlled environment rooms with HEPA filters.

Figure 1

aROCs corresponding to the association between airborne particle counting and EBC in the training period. aROC: area under the ROC curve.