Sedation Protocol During Bevacizumab Intravitreal Injection in Preterm Infants With Retinopathy of Prematurity

This study describes outcomes of intravenous (IV) analgesics and sedatives for bedside intravitreal bevacizumab injections for retinopathy of prematurity.

This retrospective study included infants receiving intravitreal bevacizumab injections between January 2012 and May 2016. Infants were excluded if bevacizumab was administered under general anesthesia or for incomplete records. Data collection included demographics, sedation and analgesia regimen, and cardiopulmonary adverse events (AEs). The primary objective was to identify the median doses of the IV analgesics and sedatives. The secondary objectives included the number of patients with cardiopulmonary AEs and those with procedure success, defined as procedure completion without interruption and absence of interventions.

Fifteen infants were included. Fourteen (93.3%) were initiated on a fentanyl infusion at a median of 2 mcg/kg/hr (IQR, 2-3.6), and 12 (80%) received midazolam infusions at a median of 0.06 mg/kg/hr. All patients received at least 1 IV neuromuscular blocker dose just prior to the procedure. Only 2 patients (13.3%) required an increase in their fentanyl or midazolam infusions. Procedure success was achieved in 13 patients (86.7%). Five patients (33.3%) experienced 1 cardiopulmonary AE. One patient (6.7%) had a delay in the procedure, and 1 patient (6.7%) required naloxone. Despite this, the procedure was completed in all patients.

Most received fentanyl and midazolam infusions with a dose of vecuronium just prior to the procedure. Thirteen (86.7%) met the criteria for procedure success. One-third experienced a cardiopulmonary AE. Future studies are needed to identify the optimal agents and route of administration for this procedure.