Questioning the legitimacy of rigid bronchoscopy as a tool for establishing the diagnosis of a bronchial foreign body.

OBJECTIVES: Rigid bronchoscopy (RB) is the principal method used for the extraction of a tracheo-bronchial foreign body (FB), but its use as a diagnostic tool implies a certain rate of negative exams, exposing the child to the risk of procedure and anesthesia-related complications. Technological progress has improved the accuracy and availability of non-invasive modalities, such as CT scan and fluoroscopy. Our aim is to review our experience in the routine use of bronchoscopy for a suspected FB aspiration, and evaluate the adequacy of our current attitude in light of these alternatives.

METHODS: We performed a retrospective review of cases where bronchoscopy was used in the management of a suspected airway FB, and analysis of the correlation between the clinical and radiological data and the bronchoscopy's results. In addition we reviewed the literature concerning the use of RB and alternative means of diagnosis such as CT scan, fluoroscopy and flexible bronchoscopy.

RESULTS: Thirty-two patients underwent bronchoscopy to rule out a FB aspiration under general anesthesia. No FB was found in 8 cases (25%). Cough and a history of choking were the most sensitive parameters (sensitivity 100% and 80% respectively), but had a low specificity. Stridor was the most specific sign (88% specificity), but was not sensitive. Chest radiography had 25% sensitivity, and 62.5% specificity. Flexible bronchoscopy changed the management in 22% of cases, sparing RB.

CONCLUSIONS: Basing the decision to perform RB solely on the clinical findings and chest radiography entails a 25% rate or more of negative exams. CT scan appears to be the most accurate non-invasive tool for ruling out the presence of a FB but its use cannot be systematic due to its complexity and the risks of exposure to radiation. Digital substraction fluoroscopy is a safe and simple mean to confirm the presence of air trapping generated by a bronchial obstruction, but it is not sensitive enough to definitively rule out a FB. We propose a stepwise approach using fluoroscopy or possibly flexible bronchoscopy under sedation, in order to reduce the number of negative RBs while restricting the use of the CT scan.