PRS AAPS Oral Proofs 2016
Jordan Swanson, MD,* Wen Xu, BS,* Tomas Aleman, MD,† Gui-Shuang Ying, PhD,† Wei Pan, MS,† Grant Liu, MD,* Shih-Shan Lang, MD,* Patrick Gerety, MD,* Gregory Heuer, MD, PhD,* Philip Storm, MD,* Scott Bartlett, MD,* William Katowitz, MD,* Jesse Taylor, MD*
From the *Children’s Hospital of Philadelphia, Philadelphia, Pa.; and †University of Pennsylvania, Philadelphia, Pa.
PURPOSE: Detecting elevated intracranial pressure (ICP) in patients with craniosynostosis may enable timely intervention to prevent neurocognitive impairment, but it is challenging with conventional methods. This study employed spectral-domain optical coherence tomography (OCT) to noninvasively quantify retinal laminae and sought to determine both the diagnostic validity of this modality and the patterns of elevated ICP among children with craniosynostosis.
METHODS: Quantitative retinal parameters were assessed using perioperative OCT in 106 subjects (age, 0.2–18 years). OCT findings in 30 patients with intracranial pathology (craniosynostosis or hydrocephalus) who also underwent direct intraoperative ICP measurement were compared with 31 age-matched negative controls. Cut-points of pathology were derived from the 95% confidence interval of negative controls and receiver operating characteristic curves. These diagnostic parameters were then applied to the overall cohort of 71 craniosynostosis patients to determine the patterns of elevated ICP.
RESULTS: ICP correlates significantly with OCT parameters: maximal retinal nerve fiber layer thickness, maximal retinal thickness, and maximal anterior retinal projection (r ≥ 0.38, P ≤ 0.04). OCT-measured retinal parameters demonstrated 93% sensitivity and 73% specificity for detecting elevated ICP. OCT parameters had high intereye agreement [intraclass correlation coefficient (ICC), 0.83–0.93] and high intragrader and intergrader agreement (ICC ≥ 0.94). When applied to all craniosynostosis subjects, 22 (47%) of 47 undergoing initial cranial vault expansion had elevated ICP. Median age at initial expansion trended toward being greater among those with elevated ICP (11.4 vs 7.8 months, P = 0.06).
CONCLUSIONS: OCT is both sensitive and specific at noninvasively detecting elevated ICP in craniosynostosis patients and may alter treatment patterns in this population.