An investigation of suitable bag materials for the collection and storage of breath samples containing hydrogen cyanide.

The SPACE study will assess exhaled breath hydrogen cyanide (HCN) concentrations as a marker of Pseudomonas aeruginosa (PA) infection in 240 children with cystic fibrosis (CF). It will use off-line selected ion flow tube mass spectrometry (SIFT-MS) analysis and so we needed to investigate which breath sampling bag material to use, the maximum storage time before analysis and the benefit of warming the bag samples. We studied 15 children with CF, 8 had chronic PA infection and 7 did not. Each exhaled directly into the instrument (on-line) and also into two 25 µm thick Nalophan (25N), two 70 µm Nalophan (70N) and two Tedlar® bags. Bags were stored at 20 or 37 °C. HCN concentrations were analysed at 1, 6, 24 and 48 h (off-line). Acetone and water vapour concentrations were also measured in parallel. Correlation between on-line and off-line concentrations measured by SIFT-MS was better for all compounds and bag types at 37 °C. The median (IQR) on-line HCN concentration was 8.9(4.4-13.7) parts per billion by volume, ppbv. Both on-line and off-line HCN concentrations were significantly higher in patients with PA infection than those without. At 37 °C the correlation between on-line and off-line HCN concentrations was good up to 6 h in the 25N bag (R(2) = 0.79) and up to 24 h for the 70N and Tedlar bags (R(2) = 0.82 and 0.86). The correlation between on- and off-line acetone concentrations at 37 °C was good up to 24 h in 25N, 70N and Tedlar bags (R(2) = 0.89, 0.93 and 0.97). In all three types of bag the water vapour concentration fell quickly and by 24 h was equivalent to that of lab air. Samples stored in Tedlar or 70N bags, warmed to 37 °C and analysed within 24 h, give HCN and acetone concentrations which correlate well with on-line measurements.