[en] Background: Asthma diagnosis is based on the presence of symptoms and the demonstration of airflow variability. Airway inflammation measured by fractional exhaled nitric oxide, measured at a flow rate of 50 ml/s (FE(NO50) ) remains a controversial diagnostic tool. Aim: To assess the ability of FE(NO50) to identify bronchial hyperresponsiveness (BHR) to methacholine (provocative concentration of methacholine causing a 20% fall in FEV(1) ; PC20M </= 16 mg/ml) and to establish whether or not symptoms relate to FE(NO50) and PC20M in patients with no demonstrated reversibility to beta(2) -agonist. Methods: We conducted a prospective study on 174 steroid naive patients with respiratory symptoms, forced expiratory volume in 1 s (FEV(1) ) >/= 70% predicted and no demonstrated reversibility to beta(2) -agonist. Patients answered to a standardised symptom questionnaire and underwent FE(NO50) and methacholine challenge. Receiver-operating characteristic (ROC) curve and logistic regression analysis assessed the relationship between PC20M and FE(NO50) , taking into account covariates (smoking, atopy, age, gender and FEV(1) ). Results: A total of 82 patients had a PC20M </= 16 mg/ml and had significantly higher FE(NO50) (19 ppb vs. 15 ppb; p < 0.05). By constructing ROC curve, we found that FE(NO50) cut-off value of 34 ppb was able to identify not only BHR with high specificity (95%) and positive predictive value (88%) but low sensitivity (35%) and negative predictive value (62%). When combining all variables into the logistic model, FE(NO50) (p = 0.0011) and FEV(1) (p < 0.0001) were independent predictors of BHR whereas age, gender, smoking and atopy had no influence. The presence of diurnal and nocturnal wheezing was associated with raised FE(NO50) (p < 0.001 and p < 0.05, respectively). Conclusion: The value of FE(NO50) > 34 ppb has high predictive value of PC20M < 16 in patients with suspected asthma in whom bronchodilating test failed to demonstrate reversibility or was not indicated. However, FE(NO50) </= 34 ppb does not rule out BHR and should prompt the clinician to ask for a methacholine challenge.