Abstract 1069

Awake endotracheal intubation is the current standard of care in most neonatal intensive care units. Little information exists regarding the potential risks or benefits of sedation for this procedure in premature infants. Objective: to assess whether sedation with Midazolam for premature infants undergoing endotracheal intubation will improve physiologic stability and thereby increase the success rate of the procedure. Design: randomized, double-blind, placebo-controlled study enrolling infants less than 37 weeks gestational age who required tracheal intubation in two Level III intensive care nurseries. Infants were randomly assigned to one of three groups: Group I (n=3) received placebo, Group II (n=6) received Atropine and placebo, and Group III (n=7) received Atropine and Midazolam. Infants could be re-randomized to a different group after completion of each intubation procedure. Heart rate, blood pressure and oxyhemoglobin saturation were obtained at 10 minute intervals for each infant. Behavioral state was assessed by Bruck scores every thirty minutes for 4 hours after the procedure. Looking for a 20% improvement in success rate with sedation, beta = 0.1 and an alpha = 0.05, power analysis revealed that an n=15 was required for each group, total n=45. Statistical analysis included the Student's t Test, Chi Square Test and the Mann-Whitney U Test. A p value less than 0.05 was considered to be significant. Results: eight patients underwent 16 intubation procedures. The study was terminated and the data reviewed early due to concerns over adverse events. Infants' mean gestational age was 26 ± 2 weeks and birthweight was 950 ± 209 grams. There was no statistical difference between the groups for postnatal age or weight at the time of the intubation procedure. There was no statistical difference in the number of attempts before successful intubation between Group I (3 ± 2) and Group II (2 ± 1, p=0.30), and Group I vs. Group III (2 ± 3 attempts, p=0.49). The number of infants with one or more oxyhemoglobin desaturation during intubation was significantly greater for those in Group III (86%) compared to those in Group I (0%, p=0.01), but not statistically different when comparing Group II (50%) to Group I (p=0.13). Cardiopulmonary resuscitation was required in 29% of Group III vs. 0% of Group I (p=0.30) and vs. 0% of Group II (p=0.15). There was no statistical difference in bradycardia, hypertension or Bruck scores between the three groups. Conclusions: premature infants receiving Midazolam are at increased risk for oxyhemoglobin desaturation during tracheal intubation. There was also a trend toward increased need for cardiopulmonary resuscitation in infants receiving Midazolam. Midazolam should not be used routinely prior to tracheal intubation until further investigation with a larger number of infants clarifies its safety and efficacy profile.