To the Editor:-Fiberoptic tracheal intubation of the infant may be assisted via a laryngeal mask airway (LMA), a standard mask, or a ventilating mask. [1] Of these devices, only the LMA acts as an oropharyngeal-laryngeal conduit, through which a flexible fiberoptic bronchoscope may be placed directly above the vocal cords. Unfortunately, the LMA is poorly tolerated by the awake infant. We describe an alternate device that facilitated fiberoptic bronchoscopic tracheal intubation of an infant with an unstable cervical spine who could not be safely anesthetized before intubation.

A 7-month-old expremature infant with a history of bronchopulmonary dysplasia, apnea and bradycardia of prematurity, and chronic respiratory failure that required prolonged intubation was admitted with rapidly progressive upper extremity weakness. A magnetic resonance imaging (MRI) examination was indicated to rule out a space-occupying lesion that involved the spinal cord. The combination of the patient's medical history and his remote position while in the MRI scanner necessitated tracheal intubation with controlled ventilation. Because of his progressive paralysis, we were compelled to assume that his cervical spine was unstable, and that direct laryngoscopy might result in permanent neurologic damage. In summary, we were confronted with a 7-month-old boy with an unstable cervical spine who could not sustain more than mild sedation for the fiberoptic placement of an endotracheal tube.

Fiberoptic bronchoscopy was performed in the operating room with an otolaryngologist standing by to perform emergency tracheostomy if necessary. After applying topical anesthesia to the child's oropharynx, using 2% lidocaine via atomizer, two attempts were made to place a 3.4-mm outer diameter bronchoscope (Pentax FB-10 X, Pentax Industries, Orangeburg, New York) into the patient's trachea. On each occasion, the child struggled and the bronchoscopy was aborted. At this point, a nipple from a baby bottle (Standard Nipple Unit, Ross Laboratories, Columbus, Ohio) was produced and a 8-mm hole was cut obliquely into the end of the nipple (Figure 1). A mark was made on the rim of the nipple in alignment with the hole, to facilitate orientation of the hole inferiorly. When introduced into the child's mouth, the nipple was eagerly accepted. Using the nipple as a conduit for the fiberoptic bronchoscope, an excellent view of the superior aspect of the epiglottis and larynx was obtained. The supraglottic region was anesthetized with 2% lidocaine injected through the auxiliary port of the bronchoscope. The bronchoscope was then advanced inferiorly past the vocal cords. A 4.0-mm internal diameter uncuffed endotracheal tube was then passed over the bronchoscope, through the nipple, and past the vocal cords. After securing the endotracheal tube, the child underwent MRI examination, which revealed a hemangioendothelioma with cervical nerve root involvement. After tumor excision, the patient had minimal residual neurologic impairment.

Rarely, the need arises to secure the airway of an awake infant without the use of a laryngoscope. We describe the use of a modified nipple as a oropharyngeal-laryngeal conduit to aid fiberoptic intubation of the trachea in this challenging context.

Randall Goskowicz, M.D., Assistant Clinical Professor of Anesthesiology, Electronic mail: rgoskowicz@ucsd.edu.

Henri G. Colt, M.D., Associate Professor of Medicine, Department of Anesthesiology, University of California, San Diego, 200 West Arbor Drive MC 8380, San Diego, California 92103.

Louis D. Voulelis, M.D., Anesthesia Associates of Boise, Staff Anesthesiologist, St. Luke's Regional Medical Center, Boise, Idaho 83702.

(Accepted for publication August 6, 1996.)