Measuring Aqueduct of Sylvius Cerebrospinal Fluid Flow in Multiple Sclerosis Using Different Software.

Maria Marcella Laganà,Dejan Jakimovski,Niels Bergsland,Robert Zivadinov,Michael G Dwyer,Francesca Baglio

doi:10.3390/diagnostics11020325

Maria Marcella Laganà, Dejan Jakimovski + Show 4 more

Open Access

https://doi.org/10.3390/diagnostics11020325

Copy DOI

Abstract

Aqueduct of Sylvius (AoS) cerebrospinal fluid flow can be quantified using phase-contrast (PC) Magnetic Resonance Imaging. The software used for AoS segmentation might affect the PC-derived measures. We analyzed AoS PC data of 30 people with multiple sclerosis and 19 normal controls using three software packages, and estimated cross-sectional area (CSA), average and highest AoS velocity (Vmean and Vmax), flow rate and volume. Our aims were to assess the repeatability and reproducibility of each PC-derived measure obtained with the various software packages, including in terms of group differentiation. All the variables had good repeatability, except the average Vmean, flow rate and volume obtained with one software package. Substantial to perfect agreement was seen when evaluating the overlap between the AoS segmentations obtained with different software packages. No variable was significantly different between software packages, with the exception of Vmean diastolic peak and CSA. Vmax diastolic peak differentiated groups, regardless of the software package. In conclusion, a clinical study should preliminarily evaluate the repeatability in order to interpret its findings. Vmax seemed to be a repeatable and reproducible measure, since the pixel with its value is usually located in the center of the AoS, and is thus unlikely be affected by ROI size.

Highlights

The cerebrospinal fluid (CSF) has various fundamental vital functions: it acts as a buffer, protecting the brain during head trauma, supports the brain weight, and helps maintaining stable central nervous system temperature [1]
We analyzed the Aqueduct of Sylvius (AoS) phase contrast (PC)-Magnetic Resonance Imaging (MRI) data of 30 people with multiple sclerosis (pwMS) and 19 normal controls (NC) using three different commercially available software packages, quantified the flow over the cardiac cycle, and estimated various measures of interest typically used in clinical studies
Our General linear models (GLM) analyses revealed that age, sex and diagnosis of MS had an effect on many variables, but that the software used did not have a significant effect on any variable

Summary

Introduction

The cerebrospinal fluid (CSF) has various fundamental vital functions: it acts as a buffer, protecting the brain during head trauma, supports the brain weight, and helps maintaining stable central nervous system temperature [1]. The CSF provides nutrition, waste removal [2], maintains stable intracranial pressure after the intracranial volumetric increment during systole [3,4]. CSF flow can be noninvasively quantified using cardiac-gated cine phase contrast (PC) Magnetic Resonance Imaging (MRI) [5]. 2D-PC, quantifying the flow passing through an acquisition plane, is routinely used for examining the Aqueduct of Sylvius (AoS) CSF hydrodynamics in people with normal pressure hydrocephalus undergoing shunt placement procedures [6], and in subjects with aqueductal stenosis [7]. AoS PC-MRI provide indirect information on intracranial pressure [8].

Objectives

Methods

Results

Conclusion