Non-Invasive Estimation of Intracranial Pressure by Magnetic Resonance Imaging Assisted Orbital Subarachnoid Space Measurement: The Beijing Intracranial and Intraocular Pressure (iCOP) Study

Abstract

Purpose: The orbital subarachnoid space surrounding the optic nerve is continuous with the circulation system for cerebrospinal fluid (CSF) and can be visualized using magnetic resonance imaging (MRI). Patients with increased ICP have a wider than normal orbital CSF space and vice versa. We hypothesized that the orbital subarachnoid space width (OSASW) is correlated with and serves as a surrogate for ICP. Our aim was to develop a method for non-invasive quantitative ICP estimation by MRI-assisted OSASW.

Methods: This clinical inter-method study included consecutive patients aged between 18 and 75 years who underwent lumbar puncture and 3.0-Tesla orbital MRI. The study population was randomly divided into a training group and a test group. The main outcome measure was OSASW at 3, 9, and 15 mm behind the globe. In training group, a non-invasive quantitative ICP assessment method was developed. In test group, the reliability and accuracy of the non-invasive ICP assessment were evaluated.

Results: Seventy-four patients were enrolled into this study. In univariate analysis, lumbar CSF-pressure was significantly associated with the OSASW at all three measurement locations (r ranging from 0.83 to 0.88; p<0.0001) within a lumbar CSF-pressure range 3.7-26.5 mm Hg, with body mass index (BMI; r=0.61; p<0.0001) and with mean arterial blood pressure (MABP; r=0.55; p<0.0001) in the training group. These associations remained statistically significant in multivariate regression. The weighting functions of non-invasive ICP prediction were: Non-invasive ICP=9.31×OSASW+0.48×BMI+0.14×MABP-19.94 (OSASW at 3 mm behind the globe); Non-invasive ICP=16.95×OSASW+0.39×BMI+0.14×MABP-20.90 (at 9 mm); and Non-invasive ICP=17.54×OSASW+0.47×BMI+0.13×MABP-21.52 (at 15 mm). In the independent test group, mean lumbar CSF-pressure (13.6 (5.1) mm Hg) did not differ significantly from mean MRI-derived calculated ICP (OSASW at 3 mm: 12.7 (4.2) mm Hg (p=0.07); at 9 mm: 13.4 (5.1) mm Hg (p=0.35); and at 15 mm: 14.0 (4.9) mm Hg (p=0.87)). The MRI-derived ICP assessment based on the OSASW at 9 mm and at 15 mm had higher reliability and greater accuracy.

Conclusions: MRI-assisted measurement of the OSASW is useful for the non-invasive quantitative estimation of the ICP, if BMI and MABP as contributing parameters are taken into account.

Keywords: 550 imaging/image analysis: clinical • 468 clinical research methodology • 613 neuro-ophthalmology: optic nerve