Todetermine the feasibility of cellular and biochemical analysis of sputum induced after inhalation of hypertonic (3%) saline, we analyzed sputum induced in 10 healthy and in 18 asthmatic subjects. We also analyzed saliva samples from all subjects. The entire sputum sample and the saliva sample were reduced using dithiotreitol, and cell counts and differentials were determined. Biochemical analysis was performed on sputum and saliva supernatants obtained after centrifugation. We found that induced sputum from asthmatic subjects had a higher percentage of eosinophils [8.1 ą 3.43 (mean ą SEM) versus 0.03 ą 0.02%, p< 0.009 J (after excluding squamous cells) and also had higher levels of albumin (232.3 ą 54.8 versus 79.5 ą 9.7 J. l9/ml, p< 0.02), fibrinogen (44.2 ą 11.6 versus 11.9 ą 2.5~ g/ml, p< 0.008) and eosinophil cationic protein (ECP)(142.6 ą 34.2 versus 26.1 ą 4.7 ng/ml, p< 0.006) but not of histamine or tryptase. In saliva, squamous cells made up more than 99% of the cells in both groups, and protein concentrations were not significantly different. We conclude that cellular and biochemical analysis of induced sputum is feasible in healthy and in asthmatic subjects and that it reveals differences similar to those reported from analyses of bronchial lavage fluid.

The analysis of cells and mediators in airway lining fluid from asthmatic subjects has traditionally been performed on samples of spontaneously expectorated sputum or on fluid obtained by bronchoscopy and lavage. Both sampling methods are limited in their applicability. Many asthmatic and healthy subjects are unable to spontaneously produce sputum, and bronchoscopy can be employed only to collect airway samples from research subjects who are well enough, and adventurous enough, to tolerate it. In addition, bronchoscopy cannot easily be applied repeatedly to follow the evolution of changes in airway lining fluid over short periods of time.