Twenty-four consecutive patients scheduled for fiberbronchoscopy were randomized to receive double-blind either intravenous (1.5 mg/kg) or laryngotracheal (3 mg/kg) lidocaine to evaluate the influence on post-bronchoscopic laryngospasm, pain in the throat and coughing. Plasma lidocaine concentrations were analyzed 5, 15, 30 and 60 min after administration. None of the patients demonstrated laryngospasm or pain in the throat during the first hour after bronchoscopy. Patients receiving topical lidocaine coughed significantly more than patients receiving intravenous lidocaine, with a median number of coughs of 20 compared to 4, during the first hour (P<0.01). The plasma lidocaine concentrations were significantly higher after intravenous than after topical administration (P<0.001). After intravenous administration the plasma lidocaine concentrations exceeded the accepted level for potential toxicity in five out of 11 patients, but none of the patients developed toxic symptoms and no side-effects were observed.