Purpose: The aim of this study was to present the entire spectrum of pediatric bronchoscopy performed for foreign body aspiration (FBA), with emphasis on accuracy of diagnostic tools, technical aspects, and predictors of complications. Methods: Records of patients who underwent bronchoscopy for evaluation of FBA in our unit from 1991 to 2000, inclusive, were reviewed retrospectively. Age, sex, history of FBA, symptoms, results of radiologic studies, bronchoscopy findings, complications, and outcome were recorded and statistically analyzed to find out the most accurate diagnostic tool. Results: A total of 740 bronchoscopies were done in 663 children (402 boys, 261 girls) presenting with a mean age of 3.1 ± 0.1 years. FBA was confirmed in 563 (85%) patients, whereas normal bronchoscopic findings, signs of pulmonary infection, and endobronchial mass were noted in 43 (6%), 54 (8%), and 3 (0.4%) patients, respectively. There was significant difference between patients with and without FBA with regard to presence of definite history (91% v 54%), normal physical examination findings (14% v 46%), and normal radiologic findings (13% v 31%). However, none of these parameters or their associations were found to be reliable to predict the presence of FBA or clinical complications. Presence of history was the most sensitive (91%), accurate (84%), and specific (46 %) diagnostic tool. Bronchoscopic removal of-foreign bodies was succeeded in 558 (99%) children. Worsening of respiratory tract infection (n = 13), cardiac arrest (n = 6), laryngeal edema (n = 5), pneumothorax (n = 5), pneumomediastinum (n = 2), tracheal laceration (n = 2), and bronchospasm (n = 2) were the life-threatening complications observed in 21 (4%) patients with FBA and 14 (14%) patients without FBA ( P < .05). There were 5 (0.8%) deaths. Of these, 2 patients presented with cardiopulmonary arrest immediately after FBA. Although foreign bodies were removed as quickly as possible, cardiac arrest was irreversible. The remaining 3 patients died of complications of bronchoscopy as irreversible cardiac arrest during bronchoscopy (n = 1) and worsening of respiratory tract infection after bronchoscopy (n = 2). Conclusions: The third year of the life carries the highest risk for FBA. There are no specific symptoms and signs to make a clear-cut differential diagnosis between FBA and respiratory tract infection. Bronchoscopy is invariably indicated on the basis of reliable history alone even when symptoms are minimal, and imaging studies are negative. Secondary bronchoscopy should be done in patients with persistent signs and symptoms to rule out overlooked organic foreign body particles or to remove persistent granulation tissue to avoid long-term complications necessitating lobectomy. The long duration of the procedure, presence of dense granulation tissue, and type of foreign body are important predictors of complications. Bronchoscopy should be regarded as an expert procedure and done with great care to avoid lethal complications. Differential diagnosis of respiratory tract infection by various diagnostic tools is of utmost important to avoid morbidity and mortality related to needless bronchoscopy.