Effect of sputum processing with dithiothreitol on the detection of inflammatory mediators in chronic bronchitis and bronchiectasis.

Sputum analysis is used increasingly to assess airway inflammation in patients with chronic obstructive pulmonary disease, including those with chronic bronchitis and bronchiectasis. However, it is not known whether dithiothreitol (DTT), a reducing mucolytic agent regularly used to homogenise sputum, affects the detection of inflammatory mediators in the sputum soluble phase from such patients. Thirty two spontaneous sputum samples were collected from 13 patients with chronic bronchitis and 17 with bronchiectasis. An aliquot from each sample was treated with either freshly prepared 0.1% DTT plus normal saline (NaCl) or NaCl alone, then ultracentrifuged to obtain the sputum sol phase. Interleukin (IL)-1beta, IL-6, IL-8, leukotriene B(4) (LTB(4)), secretory leukoprotease inhibitor (SLPI), alpha-1-antitrypsin (alpha(1)-AT), and tumour necrosis factor alpha (TNFalpha) were measured by ELISA, and neutrophil elastase (NE) and myeloperoxidase (MPO) by chromogenic substrate assay. The effect of DTT on the detection of assay standards was also determined. Median levels of IL-1beta, IL-6, IL-8, SLPI, and NE were similar in the DTT and NaCl treated samples. There was a significant reduction in median (IQR) levels of detectable TNFalpha (0.07 (0.00-0.47) pM v 0.90 (0.06-6.98) pM, p<0.001), LTB(4) (1.67 (1.31-2.64) nM v 2.29 (0.95-4.22) nM, p<0.05) and MPO (0.00 (0.00-0.00) mg/l v 4.48 (0.00-33.66) mg/l, p<0.001) and a small increase in the median alpha(1)-AT concentration (0.05 (0.03-0.08) nM v 0.03 (0.02-0.08) nM, p<0.01) in the DTT treated samples. DTT had no effect on the assay standards for IL-1beta, IL-8 or TNFalpha, but at higher concentrations it did affect IL-6, SLPI, NE, and LTB(4) standards (43%, 70%, 76% and 643% of control value for top standard, respectively) and at all concentrations DTT completely abolished MPO activity. Sputum processing with DTT significantly reduces the detectable concentration of TNFalpha, LTB(4) and MPO, and produces a small but significant increase in median alpha(1)-AT levels. To avoid this problem we recommend that an untreated aliquot of sputum be retained for cytokine analysis, unless the assay has been specifically validated.