Intranasal Dexmedetomidine for Laceration Repair in Children: a Dose-finding Study

Design: This study will be designed as a phase II single-arm dose-escalation pilot study using the continual reassessment method.

Protocol: Sedation will be measured using the Pediatric Sedation State Scale (PSSS), an instrument validated for video scoring of children undergoing painful procedures. The PSSS is scored from 0 (dangerously sedated) to 5 (least sedated) easily by non-medical personnel and assesses over sedation and under sedation. Adequate sedation is a score of 2 or 3. Participants will be consecutively assigned to increasing doses of IND from 1-4 mcg/kg based on the continual reassessment method. Participants will be assigned to doses of IND from 1-4 mcg/kg, increasing in whole number increments. Initially three participants will receive 1mcg/kg and the number of participants at each dose of IND will be recorded. Data from these participants will be used to update a Bayesian model for the dose-response curve for all three categories of sedation. The following three participants will be assigned the dose with the highest posterior probability of an efficacy close to 0.8. This balances the need to determine the most efficacious dose but prevents an excessive number of over-sedations. For the Bayesian dose response model, a determination will be made as to the overall score category for each participant ("adequate", "over", or "under sedated" based on the PSSS). To be scored as "adequate", a participant must have a PSSS score of 2 or 3 for at least 90% of observations from initial positioning to tying of the last suture. If a participant does not retain a PSSS score of 2 or 3 for at least 90% of the observations, they will be categorized as either over or under-sedated, if the majority of the remaining PSSS scores are 0 or 1 or 4 or 5, respectively. Furthermore, if the participant remains awake, but not distressed during the procedure, they will be scored as a 2 based on the PSSS. However, for the purposes of the Bayesian dose response model, they will be scored as "under sedated" to avoid concluding that a lower dose of IND is effective based on the outcomes for participants that did not require sedation. Finally, participants who are noncompliant with IND will be categorized as an over-sedation as it is assumed that this dose was not well tolerated by the participant and dose escalation should be avoided. Data will be reviewed after each dose increment by a data safety monitoring board, who will confirm it is safe to escalate. Permissible co-interventions include topical and subcutaneous anesthetic, oral or IV analgesics, non-pharmacologic strategies for pain and distress.

Screening and Enrollment: Participants will be consecutively screened for eligibility during the hours of study recruitment.

Sample Size: This was calculated using the Bayesian continual reassessment method (20). Based on an adverse event rate of 10%, 4 dosing levels, an odds ratio for the effect size of 1.8, and a phase II trial accuracy level of 60%, an estimated a sample size of 55 participants was calculated.