In Vitro Study of Cerebrospinal Fluid Dynamics in a Shaken Basal Cistern after Experimental Subarachnoid Hemorrhage

et al. (2012) In Vitro Study of Cerebrospinal Fluid Dynamics in a Shaken Basal Cistern after Experimental Subarachnoid Hemorrhage. PLoS ONE 7(8): e41677. doi:10.1371/journal.pone.0041677 In Vitro Study of Cerebrospinal Fluid Dynamics in a Shaken Basal Cistern after Experimental Subarachnoid Hemorrhage Andreas Spuler 0 Ulrich Kertzscher 0 Torsten Schneider 0 Leonid Goubergrits 0 Klaus Affeld 0 Daniel Ha nggi 0 Robert Loch Macdonald, St Michael's Hospital, University of Toronto, Canada 0 1 Biofluid Mechanics Laboratory, Charite - Universita tsmedizin Berlin , Berlin, Germany , 2 Department of Neurosurgery, Heinrich-Heine-University , Du sseldorf, Germany , 3 Department of Neurosurgery , Helios Klinikum Berlin-Buch, Berlin , Germany Background: Cerebral arterial vasospasm leads to delayed cerebral ischemia and constitutes the major delayed complication following aneurysmal subarachnoid hemorrhage. Cerebral vasospasm can be reduced by increased blood clearance from the subarachnoid space. Clinical pilot studies allow the hypothesis that the clearance of subarachnoid blood is facilitated by means of head shaking. A major obstacle for meaningful clinical studies is the lack of data on appropriate parameters of head shaking. Our in vitro study aims to provide these essential parameters. Methodology/Principal Findings: A model of the basal cerebral cistern was derived from human magnetic resonance imaging data. Subarachnoid hemorrhage was simulated by addition of dyed experimental blood to transparent experimental cerebrospinal fluid (CSF) filling the model of the basal cerebral cistern. Effects of various head positions and head motion settings (shaking angle amplitudes and shaking frequencies) on blood clearance were investigated using the quantitative dye washout method. Blood washout can be divided into two phases: Blood/CSF mixing and clearance. The major effect of shaking consists in better mixing of blood and CSF thereby increasing clearance rate. Without shaking, blood/CSF mixing and blood clearance in the basal cerebral cistern are hampered by differences in density and viscosity of blood and CSF. Blood clearance increases with decreased shaking frequency and with increased shaking angle amplitude. Head shaking facilitates clearance by varying the direction of gravitational force. Conclusions/Significance: From this in vitro study can be inferred that patient or head shaking with large shaking angles at low frequency is a promising therapeutic strategy to increase blood clearance from the subarachnoid space. - Despite the elaborate current treatment strategies, cerebral arterial vasospasm leading to delayed cerebral ischemia constitutes the major delayed complication following aneurysmal subarachnoid hemorrhage (SAH). Vasospasm is associated with significant morbidity and mortality rates [1]. Independent predictors of cerebral vasospasm are clot volume in the subarachnoid space [2,3] and clot clearance rate [4]. Being a function of the amount of subarachnoid blood, the incidence of vasospasm can be reduced by increasing blood clearance from the basal cerebral cisterns. This can be achieved by means of increased perfusion of the subarachnoid space with CSF alone [5] or combined with intracisternal thrombolysis [6]. Clearance of cisternal blood may also be facilitated by repeated head motion. Two different head motion strategies, translational movement [7,8] and bidirectional rotation [9], have been applied clinically. Since in these studies head motion was combined with an intensified lumboventricular lavage [9], the contribution of bidirectional rotation to the clearance of the subarachnoid space is difficult to dissect from the effect of increased CSF flow. Furthermore, none of the studies of head motion investigated variations of head shaking parameters. The aim of our study was to define clinically applicable settings for effective head shaking. Using a quantified dye washout technique in vitro, we visualized and analyzed CSF flow and cisternal blood clearance in the central basal cistern. In order to understand the basic blood washing process in the basal cistern, we investigated the clearance of a blood model during head shaking with a steady net CSF flow. Materials and Methods Phantom Model of the Basal Cistern The study complies with the declaration of Helsinki. The study was done in a model constructed from human magnetic resonance imaging data of the brain. All data are from one single human who is co-author (DH) and gave written voluntary informed consent. Approval was given by the ethic committee of the Heinrich-HeineUniversity Du sseldorf, Germany. The model of the basal subarachnoid cistern consists of the prepontine cistern, the pontocerebellar cistern, and the pentagonal cistern. It was fabricated using a stereolithography (STL) geometry. Magnetic resonance imaging (MRI) data were acquired from the above mentioned healthy 35-year-old male volunteer using a 1.5 T scanner (Magnetom AvantoTM, Siemens, Erlangen, Germany) by means of a spin echo (SE) sequence (sequence variant SK) with a slice thickness of 2.4 mm, a repetition time (TR) of 4,000.0 ms, and an echoing time (TE) of 250.0 ms. Data acquisition resulted in a stack of 180 slices with a voxel resolution of 0.82 mm60.82 mm61.2 mm. Three-dimensional segmentation of the subarachnoid cistern was done using the software MimicsTM (Materialise NV, Leuven, Belgium) (Figure 1). Segmentation was performed manually based on boundaries defined primarily by automatic segmentation using intensity signal thresholding. Finally, the reconstructed geometry was converted into a STL file format. This geometry served as physical model in our experimental study (Figure 2A and B). The volume of the experimental basal cistern was 20 ml. Sizes of the cistern (length and width) are indicated in Figure 2B. Mean depth of the cistern was approximately 10 mm. From the reconstructed surface geometry (STL file) a 1:1 wax cast was fabricated by means of a rapid prototyping printer thermojet (PORTEC, Zella-Mehlis, Germany) with a spatial resolution of 0.1 mm (Figure 2C). Thereafter, a transparent silicone phantom model of the basal cistern was generated as block cast. For this, the wax model with ducts representing the CSF inflow and outflow tracts was embedded in a mixture of ElastosilTM RT 601 A and B (Wacker Chemie, Munich, Germany). Inflow and outflow tracts were mounted along the longitudinal axis of the cistern model at the narrowest regions of the boundaries of the reconstructed basal cistern geometry. Inside diameters of inflow and outflow tracts were 4 mm each. The diameters were selected according to the cross-sectional areas at the respective regions (basal cistern inflow/outflow) identified on the MRI data. The wax was melted out in a furnace at 150uC resulting in the transparent silicone model of the basal cistern (Figure 2D). Experimental Setup The study was performed in a custom-built experimental se (...truncated)