Endogenously released nitric oxide (NO) has been detected in the exhaled air of humans. Exhaled NO (NOexh) levels have been significantly increased in patients with inflammatory airways disorders such as asthma, and NOexh has been suggested to be a usable marker of airway inflammation. In the present study, NOexh levels were measured both in steroid-treated and untreated subjects with mild asthma, and were correlated with the degree of airway hyperresponsiveness (AHR), measured as the dose of histamine that produced a 20% decrease in FEV1 (PC(20)histamine). NOexh levels, which were significantly increased in steroid-naive patients (Group Al: NOexh = 21 +/- 11 ppb; n = 56) in comparison with levels in control subjects (Group B: NOexh = 10 +/- 2 ppb; n = 20; p < 0.001), correlated significantly with the PC(20)histamine (r = -0.65; p < 0.0001). The NOexh level was significantly lower in patients with chronic cough of other causes than bronchial asthma (Group A2: NOexh = 11 +/- 3 ppb; n = 18) when compared with the level in subjects with mild asthma (Group Al: p < 0.001). Therefore, the noninvasive measurement of NOexh allowed us to discriminate, among patients with respiratory complaints, between those with and without AHR. In asthmatic subjects treated with inhaled steroids, the NOexh levels were significantly lower (Group A3: NOexh = 13 +/- 5 ppb; n = 25) than in untreated subjects (Group Al; p < 0.01), and there was no relationship with the PC(20)histamine (r = -0.18, p = NS). These findings confirm that NOexh reflects AHR in patients with mild asthma who have not already been treated with inhaled steroids. Patients treated with inhaled steroids had an NOexh level comparable to levels in control subjects, although AHR could still be demonstrated.