Comparing invasive with MRI-derived intracranial pressure measurements in healthy elderly and brain trauma cases: A pilot study

Background: Intracranial pressure (ICP) is an important physiological parameter in several neurological disorders. Considerable effort has been made to measure ICP noninvasively. MR-based ICP (MR-ICP) is a nonempirical method based on principles of cerebrospinal fluid (CSF) physiology, where ICP is obtained from measurements of blood and CSF flows to and from the cranium during the cardiac cycle. Purpose: To compare MR-ICP with invasive ICP measurements obtained using lumbar puncture (LP) or external ventricular drainage (EVD). Study Type: Prospective, cross-sectional, observational study. Subjects: Ten cognitively healthy elderly subjects (age 69.6 ± 6.6 years; seven females) and six brain trauma patients (age 36.8 ± 19.7 years; two females). Field Strength: Velocity encoding cine phase-contrast at 1.5 T and 3 T. Assessment: MR-ICP and craniospinal compliance distribution were estimated from arterial inflow and venous outflow to and from cranium, and craniospinal CSF flow at the upper cervical region, measured using cine phase contrast MRI. LP (done 177 ± 163 days after scan) and EVD measurements (at the time of scan) were performed in lateral recumbent and supine positions, respectively. Statistical Tests: Linear regression was used to assess the relationships of MR-ICP with invasive ICP, and the dependency of these measurements on age, weight, height, and BMI. A Shapiro–Wilks test and Bland–Altman plot were respectively used to evaluate the normality and agreement between these two pressure distributions. Student's t-test was used throughout the analysis to compare differences between the EVD and LP cohorts. Results: In the combined cohort, MR-ICP and invasive ICP were positively correlated (r = 0.95, P < 0.001), with invasive ICP being higher than MR-ICP by 2.2 mmHg on average. In the healthy cohort, the cranial contribution to total craniospinal compliance was negatively correlated with MR-ICP (r = −0.90, P < 0.001). Data Conclusion: MR-ICP provides a reliable estimate of ICP, with 14 out of 16 datapoints within the clinically acceptable error. Craniospinal compliance distribution plays a role in modulating ICP in supine position. Level of Evidence: 3. Technical Efficacy: Stage 2. J. Magn. Reson. Imaging 2019;50:975–981.