Sevoflurane sedation in infants undergoing MRI: a preliminary report

BACKGROUND: Stillness during natural sleep after feeding may not be sufficient for successful magnetic resonance imaging (MRI) in small infants less than 5 kg. Sedation, using an oral agent, is often successful although the timing and depth of sedation is variable. In contrast anesthesia is always effective but is invasive and is associated with postanesthesia apnea and bradycardia in preterm and ex-preterm infants. We are developing an alternative technique involving insufflation of sevoflurane and present our initial experience. METHODS: Infants presenting for MRI were sedated by nasal insufflation of sevoflurane carried by 2 l.min(-1) oxygen. We recorded the sevoflurane administered, timing of sedation and scanning, conscious level, oxygen saturations, and recovery profile. RESULTS: Of the 13 infants studied (median postconceptional age: 46 weeks, range: 40-70 weeks; median weight: 4.4 kg, range: 3.3-6.5 kg), sevoflurane caused sleep and enabled successful imaging in 12. Six infants fell asleep within 10 min and the median maximum sevoflurane vaporizer setting for successful sedation was 4% (range: 4-8%). Before scanning, 10 infants remained easily roused by touch and two became unresponsive; one desaturated to 85% and required repositioning of the head to maintain a clear airway. Immediately after scanning all infants were easily roused by touch. CONCLUSIONS: Sedation by insufflation of sevoflurane in small infants is a simple and practical alternative technique for painless imaging such as MRI; further experience is necessary to determine its limitations.