Comparison between three transmucosal routes of administration of midazolam in children

Midazolam was applied transmucosally in 47 children randomly assigned to three different groups. Group N received 0.2 mg.kg-1 nasally, group R 0.5 mg.kg-1 rectally, and group S 0.2 mg.kg-1 sublingually. All groups were treated 60 min prior to a planned i.v. puncture with EMLA. Reliable and valid psychological parameters (such as emotional situation, shivering, awareness, respiratory rate and facial colour) were scored after premedication and before and after i.v. puncture, 20 min after premedication and until induction. A blood sample was drawn 10, 30 and 60 min after premedication and the levels of midazolam, alpha-hydroxy-midazolam, ACTH, glucose and cortisol were measured. In all three groups the plasma levels of midazolam 10 min after premedication were higher than 70 ng.ml-1 (accepted as a sedative level). 30 min after premedication the midazolam level in the sublingual group was statistically significantly higher than in the nasal group and the psychological parameters in all three groups were significantly changed (10 min after premedication). The psychological parameters were not significantly different between the three groups over the whole study. Sublingual premedication has some advantages (most readily accepted, highest plasma levels and lowest deviations) and could be the first choice in premedication of children. All three transmucosal applications are safe and well accepted, although nasal application was rejected by two of the children.