LAUSR

Pediatric Critical Care Medicine www.pccmjournal.org 285 performed 50 supervised retinal vessel ultrasounds in preparation for this study. Additional studies examining interobserver variation would be needed to determine its reproducibility. Similar to TCD, retinal ultrasound may find limited application in cases with open craniocerebral trauma or decompressive craniotomy, where regionally limited circulation can occur. There were no patients less than 30 days of age in this study. Ancillary studies for the newborn are less sensitive when determining brain death in this age group (1, 2). Finding more sensitive ancillary studies, especially in the newborn period, reducing variability and standardizing how brain death is determined are essential to avoid diagnostic error. It remains to be seen whether limitations that affect TCD interpretation in open craniocerebral trauma in adults also apply to retinal ultrasound in newborns or infants with an open fontanelle. CBF, electroencephalogram, and four-vessel angiography are well-established gold standards for ancillary studies and are performed under specific technical protocols. Additional studies are needed to determine the accuracy and reliability of ultrasound of the central retinal artery as a marker for CBF compared with the existing validated studies before ultrasound of the central retinal artery can be considered an ancillary study to assist with determination of neurologic death. Specific protocols need to be established including training and certification required to perform this type of procedure. Ultrasound of the central retinal artery has potential to become a useful screening tool to determine CBF. Whether this novel method of evaluating CBF will prove to be useful as an ancillary study supporting the determination of brain death remains to be seen.