Prediction of Intracranial Pressure From Noninvasive Transocular Venous and Arterial Hemodynamic Measurements: A Pilot Study

Abstract

Continuous measurement of intracranial pressure (ICP) requires the invasive placement of epidural, parenchymal, or intraventricular devices. For critical single-point assessments, lumbar puncture may not always be practical. An accurate, reliable, portable and noninvasive method to estimate absolute ICP remains an elusive goal. The arteries that perfuse and the vein that drains the orbit are exposed to the ambient ICP while coursing through the cerebrospinal fluid or optic nerve. The venous outflow pressure (VOP) of the central retinal vein was measured using occlusion in six intensive care patients treated for acute hydrocephalus or brain hemorrhage and in whom transducers of intracranial pressure could provide standardized continuous output. A novel adaptation of the Balliart ophthalmodynamometer was developed for use. Simultaneously, the central retinal (CRA) and ophthalmic (OA) arterial flow velocities were recorded using color Doppler imaging technique. Repeat noninvasive measurements were performed at various ICPs, (n=22 independently collected observations). Linear regression and correlation testing were performed to evaluate these variables for ICP predictability. The VOP increased linearly with ICP (r=0.87). The arterial pulsatility indices for both OA and CRA decreased inversely with ICP (r=0.66). An empiric index combining both venous and arterial parameters (VOP/Gosling Pulsatility Index [GPI]) was significantly more correlated with absolute ICP than either parameter alone (r=0.95, p<0.005, ICP=0.29+0.74 [VOP/GPI(OA)]). The feasibility to estimate ICP from transocular sonographic and dynamometric data is suggested by these preliminary data. Retinal arterial properties are important in modeling the effect of ICP on the venous outflow pressure. Our pilot results serve as a basis on which to conduct a larger prospective and blinded study.