Tracheotomy in very low birth weight neonates: indications and outcomes.

OBJECTIVE/HYPOTHESIS: To review incidence of, indications for, and outcomes of tracheotomy in very low birth weight (VLBW) infants. STUDY
DESIGN: Retrospective review in tertiary care hospital.
METHODS: Eighteen VLBW (<1,500 g) infants with bronchopulmonary dysplasia undergoing tracheotomy in the neonatal intensive care unit between October 1997 and June 2002 were studied. Controls consisted of 36 VLBW infants undergoing intubation without tracheotomy, two per study infant, matched by gestational age and weight. Outcome measures included duration and number of intubation events, time to decannulation, complications, comorbidities, length of stay, and speech, language, and swallowing measures.
RESULTS: Infants undergoing tracheotomy had an average duration of intubation of 128.8 days with a median number of 11.5 intubation events, both significantly greater than those of controls. Percentage of those with laryngotracheal stenosis was 44% of study infants had laryngotracheal stenosis compared to 1.6% in all intubated VLBW infants. The tracheotomy group had a significantly higher incidence of gastroesophageal reflux, pulmonary hypertension, and gastrostomy tube placement. The overall tracheotomy-related complication rate was 38.9%. Three were lost to follow-up, and five deaths occurred, two possibly tracheotomy-related. Six of ten were decannulated by an average time of 3.8 years, two of six after laryngotracheal reconstruction. Four of ten remained cannulated for a variety of reasons. Disorders of speech, language, and swallowing were common.
CONCLUSIONS: When considering tracheotomy in VLBW infants, the total number of intubation events should be monitored as well as the total duration of intubation. The relatively high incidence of laryngotracheal stenosis argues for earlier endoscopy and possibly earlier tracheotomy in infants with developing stenoses.