Abstract 1108 Poster Session I, Saturday, 5/1 (poster 113)

Sensorineural hearing loss (SNHL) is a significant neurologic morbidity in survivors of neonatal congenital diaphragmatic hernia (CDH), with a reported incidence up to 60%.

Methods: In a historical cohort study of 37 neonates with CDH from 1989 to 1995, we investigated the use of pancuronium bromide and common ototoxic drugs (aminoglycosides, vancomycin and loop diuretics) during the neonatal period and their relationship to SNHL detected in long term survivors.

Results: SNHL was diagnosed in 23 (62%) of 37 children with CDH, 13 (65%) of 20 ECMO-treated, and 10 (59%) of 17 conventionally treated neonates. Seventeen of the 23 SNHL had normal or mild/moderate low frequency sloping to moderate/severe high frequency loss, whereas 6 had severe/profound loss at 500 Hz and above (flat loss). There was no significant difference in demographic and neonatal characteristics, including oxygenation and ventilation variables, between the survivors with SNHL and those without SNHL. The ECMO-treated neonates had significantly lower Apgar scores, higher mean airway pressure, higher maximum oxygenation indices, lower PaO2, and longer duration of mechanical ventilation and hospitalization than conservatively treated neonates (two-way ANOVA). Survivors with SNHL had a significantly higher cumulative dose of pancuronium bromide used during the neonatal illness than those survivors without SNHL. A significant dose-dependent effect of pancuronium bromide on the type of SNHL was observed: the cumulative dose and duration of pancuronium bromide were greater in survivors with flat loss than those of high frequency SNHL, which were in turn greater than those of normal hearing (P<0.0001, one-way ANOVA). The cumulative dose and duration of pancuronium bromide significantly correlated (r=0.66-0.81) and independently predicted (adjusted r2=0.42-0.64) the greatest intensity (dB) and the widest band (lowest frequency) (Hz) of loss of SNHL. In analyzing the cumulative dose and duration of therapy for aminoglycosides, vancomycin and furosemide, no differences were identified between survivors with and without SNHL, although survivors with SNHL had a modestly higher cumulative dose of ethacrynic acid than that of survivors without SNHL. A duration of pancuronium bromide use of ≥7 days was significantly associated with SNHL in the survivors, giving a sensitivity of 57%, specificity of 86%, and positive predictive value of 68% (Chi-Square test).

Conclusion: While we demonstrate that prolonged administration of pancuronium bromide during the neonatal period is associated with SNHL in childhood survivors of CDH, further multi-centered studies are required to investigate the possible etiologies of SNHL in the high-risk population.