Abstracts from Hydrocephalus 2019: The Eleventh Meeting of the International Society for Hydrocephalus and Cerebrospinal Fluid Disorders

s from Hydrocephalus 2019: The Eleventh Meeting of the International Society for Hydrocephalus and Cerebrospinal Fluid Disorders Vancouver, Canada. 13–16 September 2019 Published: 5 December 2019 © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. I1 Thomas J. Zwimpfer MD, PhD, President, Hydrocephalus 2019 Department of Surgery, University of British Columbia, Vancouver, Canada Correspondence: Thomas J. Zwimpfer thomas.zwimpfer@ubc.ca It has been a great pleasure to host the 11th Meeting of the Hydro‐ cephalus Society (International Society for Hydrocephalus and Cerebrospinal Fluid Disorders) in Vancouver, British Columbia, Canada from September 13 to 16, 2019. This year we hosted 216 individuals from 19 countries. This annual meeting brought together individuals with a wide variety of backgrounds, including clinicians and basic scientists, varying from professors to trainees, as well as nurses, nurse-practitioners, therapists, patient advocates and volunteers. They all shared common interests and goals: to better understand the normal physiology of cerebrospinal fluid and intracranial pressure, to improve the diagnosis and treatment of hydrocephalus and other CSF disorders in both children and adults. The Pre-Meeting Educational Day focused on broad topics such as: (1) Physiology of CSF/ICP and Pathophysiology of Hydrocephalus; (2) Experimental Hydrocephalus; (3) Pediatric Hydrocephalus and; (4) Chronic Hydrocephalus, a cause of reversible cognitive impairment. Highlights of the Meeting included; •• • 36 high quality submissions to the Young Investigator Award competition. •• • Two Hakim lectures: (1) Interactions between iNPH and AD and; (2) Factors related to Ventriculomegaly and its relationship to Aging. •• • The Marmarou Lecture: The Umea Experience in CSF Dynamics in Hydrocephalus, •• • Presentation by the New International iNPH Guidelines Working group. •• • •Keynote lectures on: (1) Results of the UK Basics trial on antibiotic-impregnated shunt catheters; (2) Basic science research on shunt blockage; (3) Research on shunt outcomes; (4) Diffusion Tensor Imaging in Pediatric Hydrocephalus; (5) Artificial Intelligence in Hydrocephalus Research and; (6) Update on Models of Experimental Hydrocephalus. •• • Panel Presentation on the various International Hydrocephalus Registries. • • International Hydrocephalus Imaging Working Group Presentations. The highlight of the Social activities was a cable car trip to the top of Grouse Mountain for fine city views and dining, to experience a Lumberjack Show and a very close-up visit with two 1000 lb Grizzly bears, Grinder and Coola, who have called Grouse Mountain home for the past 18 years after being orphaned as cubs. I look forward to seeing many of you in Gothenburg, Sweden in September 2020 for the 12th Meeting of the Hydrocephalus Society. I wish you a productive and exciting year.


Highlights of the Meeting included;
•• • 36 high quality submissions to the Young Investigator Award competition. •• • Two Hakim lectures: (1) Interactions between iNPH and AD and; (2) Factors related to Ventriculomegaly and its relationship to Aging. The highlight of the Social activities was a cable car trip to the top of Grouse Mountain for fine city views and dining, to experience a Lumberjack Show and a very close-up visit with two 1000 lb Grizzly bears, Grinder and Coola, who have called Grouse Mountain home for the past 18 years after being orphaned as cubs. I look forward to seeing many of you in Gothenburg, Sweden in September 2020 for the 12th Meeting of the Hydrocephalus Society.
I wish you a productive and exciting year.

Tom Zwimpfer
Introduction: Sedation is often employed in order to improve the experience of patients who undergo lumbar punctures. The effect of sedation on lumbar puncture opening pressures is not entirely known. In this study we investigated the effect of sedation on intracranial opening pressure in a cohort of patients who underwent continuous intraparenchymal ICP monitoring. Methods: Observational study. Intraoperative intracranial opening pressures of patients undergoing ICP monitoring under sedation was compared to median 24 h ICP results and mean ICP obtained in different body positions (including lumbar puncture position). In a subset of patients, the level of sedation and CO 2 were also monitored continuously.

Introduction:
The management of acute hydrocephalus is a constantly developing and improving area within neurosurgery. We recently demonstrated a standardised perioperative care bundle and simulation training improves placement and infection rates of tunnelled external ventricular drains (EVDs)-we have since developed our practice further to include minimally-invasive bolt EVDs. These are inserted with a smaller calibre hand drill, and can be inserted in an ITU environment. This study compares the infection rates, placement accuracy, and time from decision to CSF access between the modalities. Furthermore, we have begun to introduce the use of the LiquoGuard drainage system instead of the standard gravitational Becker drain, particularly in cases with significant intraventricular blood, or where lead clinicians feel there is a high risk of drain blockage. Methods: A combined retrospective and prospective cohort study of every EVD at our quarternary referral neurosurgical centre, 1/12/18-30/4/19. Results: In the 5 months, 46 EVDs were inserted-23 of each type. Of these, 7 were connected to a LiquoGuard. Preliminary data demonstrates average time to CSF access was 140 min for bolt EVDs, and 337 min for tunnelled (p = 0.0015). Accuracy of placement was comparable between the modalities at approximately 90%. There were no infections noted. Average length of drainage was 9.7 days in bolt EVDs, and 8 days in tunnelled. Conclusions: Our data demonstrates that bolt EVDs provide a statistically significant faster time to access of the CSF, have a comparably low infection rate, and are accurately placed in trained hands as often as tunnelled EVDs.
Introduction: CT imaging remains the most common initial diagnostic modality for hydrocephalus in the United Kingdom. We explored whether the diagnosis of hydrocephalus can be achieved by non-human diagnostic agents. Several machine learning models exist for image classification, but their adoption in clinical practice for hydrocephalus in the UK remains underutilised. This study attempted to use a machine learning framework engine to classify CT images of hydrocephalus patients. Methods: A systematic image literature search through Google scholar and Google image analysis was performed to acquire suitable public domain images, while avoiding the geographic impracticalities and differences that surround patient consent and the General Data Protection Regulations for clinical image data. Selected training and test images were verified by specialist neurosurgeons, who also reviewed normal control brain images. A hierarchical supervised learning algorithm was implemented for axial static radiographs at lateral ventricular, third ventricular and cisternal level. The model was evaluated by introducing newer images, which it had not been presented with before. Results: 77.78% accuracy of recognition of hydrocephalus was achieved through training the algorithm on the small number of CT images identified in the search online (N = 20 hydrocephalus and N = 12-control images). Conclusions: Non-human diagnostic agents can achieve relatively good predictive accuracy for hydrocephalus with very little training data. Further training with more images to improve the accuracy and sensitivity of detection is currently underway. Conclusions: Identifying scores that fall beyond the normal range of measurement error is essential for assessing statistically significant change. The RCIs presented in this paper allow for clinicians to make evidence-based decisions while treating iNPH.
Introduction: While gait has been investigated using 3-D motion analysis pre and post shunt surgery, these gait parameters have not been compared to healthy persons. The objective was to characterize gait in patients with iNPH before and after VP shunt placement and compare to an unimpaired population. Methods: Seven patients with a diagnosis of iNPH were analysed preoperatively and 4 weeks postoperatively. Patients walked barefoot and unassisted on a 10 m walkway. Kinematic data from a 10 camera system and kinetic data from five force plates embedded in the walkway were processed in Visual3D, and compared to laboratory normative data from 20 healthy young adults using t-tests. Results: Preoperative temporal distance parameters were significantly different from healthy adults, and increased significantly following VP shunting, though did not reach values of healthy adults, and was much greater than the effect due to aging. Total sagittal plane range of motion (ROM) in the hip, knee and ankle showed decreased ROM compared to healthy adults (p < 0.05), and significant increases in hip and knee ROM were observed postoperatively again short of normal values. Ankle dorsiflexion and hip extension increased post shunt surgery, with maximum knee flexion showing a significant increase (p < 0.05). Conclusions: Gait was objectively quantified pre and post VP shunt placement in patients with iNPH, with all measures trending toward normative values post surgery. Further studies comparing gait patterns pre and post VP shunt placement can provide insight into the efficacy of the surgical treatment and aide in guiding clinical practice.

O18
A predictive classification for post-traumatic hydrocephalus following decompressive craniectomy for acute subdural haematoma: a london major trauma centre experience Amin Elyas 1 , Hasan Asif 1 , Curtis Offiah 2 , Chris Uff 1 Introduction: Post-traumatic hydrocephalus (PTH) is a known complication of acute subdural haematoma (ASDH) that has been managed with decompressive craniectomy (DC). The management usually requires permanent CSF diversion which in itself carries a risk of morbidity. The aetiology of PTH is unclear and has been suggested to be due to CSF flow disturbance secondary to the craniectomy but there is no correlation between DC size and incidence of PTH. We suggest a novel mechanism and predictive score for PTH in patients who have undergone DC for ASDH. Methods: Retrospective review of prospectively maintained database of patients undergoing DC for ASDH between October 2016 and April 2019. Pre-DC computerised tomography imaging was used to assess for obliteration of four CSF spaces: right Sylvian fissure, left Sylvian fissure, right cerebral convexity and left cerebral convexity. Post-operative interval CT imaging was examined to assess for incidence of PTH. Results: Fifty-eight patients (38M:20F) mean age 39.7 (± SD 15.2) underwent unilateral DC for ASDH, of these 38 (69.0%) went onto develop PTH. Our test cut off of "obliteration of 2 or more CSF spaces" Fluids Barriers CNS 2019, 16(Suppl 3):36 was able to predict PTH with sensitivity of 97.5% (95% CI: 86.8-99.9) and a specificity of 50.0% (95% CI: 26.0-74.0) with 81.25% PPV and 90% NPV. Area under ROC curve for "obliteration of 2 or more CSF spaces" predicting PTH was 0.77 (P < 0.001).
Conclusions: Early identification of radiological features of acute or impending PTH may allow for appropriate and timely CSF diversion facilitating reduction of morbidity, early discharge, cranioplasty and rehabilitation. selected. For all these cases, preoperative imaging, intraoperative recordings, and postoperative imaging were reviewed. Preoperative clinical data were compared with postoperative outcomes. Results: With the use of the flexible scope, we were able to completely aspirate intraventricular clots in patients affected by tetraventricular hemorrhage. We could effectively manage arachnoid cysts in the fourth ventricle and even in the cisterna magna from a precoronal paramedian burr hole. It was also possible to detect membranous obstructions of the cerebral aqueduct, treating hydrocephalus with endoscopic perforation of the membranes. Conclusions: Despite the lower image quality compared to the rigid scope, and the lack of dedicated instrumentation, only the flexible scope allows complete navigation of the cerebral aqueduct and fourth ventricle for cyst fenestration or complete aspiration of intraventricular hemorrhage, using a single burr hole access. A non-stenotic aqueduct can be safely navigated by a well-experienced neuroendoscopist. In our experience, there is no risk of damage to the fornix using a flexible scope.

O19
Introduction: The amount of intraventricular blood is a strong negative prognostic predictor on outcome. Therefore, massive intraventricular hemorrhages (IVHs) require aggressive and rapid management to decrease intracranial hypertension. Flexible neuroendoscopy can be used for intraventricular clots removal, allowing for rapid reduction of intracranial pressure and early removal of external ventricular drainage. Methods: We present the series of 22 patients who were treated to remove IVH at our Institution. Neuroendoscopy is indicated when IVH causes hydrocephalus, brainstem compression, and ultimately intracranial hypertension. If aneurysm is the primary cause of IVH, it must be secured before proceeding with neuroendoscopic removal of intraventricular clots. Results: All ventricles could be explored and cleared from clots, in order to restore patency of CSF pathways. The external ventricular drainage (EVD) was always left in the ventricle after surgery, although in most of the cases early removal was possible. The length of stay in ICU was reduced compared to patients who were treated only with EVD. Conclusions: Early neuroendoscopic removal of blood casting from the lateral to the fourth ventricle is a feasible approach, allowing in most instances the rapid improvement of the IVH, the decrease of EVD dependency, and shorter ICU stay. anaesthesia for bolt removal, comparing efficacy and safety against current practise. Methods: Prospective case cohort of patients undergoing removals of diagnostic ICP bolts between June 2017 and April 2019. Two groups were identified: "A" receiving oral analgesia only and "B" receiving ipsilateral supraorbital and supratrochlear nerve blocks with 5 ml of 1% lidocaine. Subjective outcomes were collected by review of patientcompleted questionnaires with white space, yes-no and 5-point Likert scale questions. Results: Eighty-five patients were fitted with ICP monitoring bolts (32M:53F, mean age 42.7 ± 16.0). Fifty-four were removed with oral analgesia only (group A) and 31 were removed with oral and regional anaesthesia (group B). Overall removal experience was 3/5 for group A and 4/5 for group B (p < 0.01). Thirty-six (70.6%) patients would have preferred a scalp block in group A. In group A, patients reported the best part of removal was "having it out" and worst was "pain and slowness". In group B, the best part of removal was that "it was painless" and worst part was during anaesthetic infiltration. Conclusions: Regional nerve blocks to the ipsilateral supraorbital and supratrochlear nerves are a safe and effective adjuvant for the painless removal of frontal ICP monitoring bolts. Introduction: Normal pressure hydrocephalus (NPH) is a critical brain disorder with gait failure, cognition impairment and urinary incontinence as its core symptoms. The high morbidity and mortality in older patients lead to a heavy economic and social burden. The diagnosis of NPH, especially the idiopathic normal pressure hydrocephalus (iNPH) is a challenge for the diversity and of coexistence of symptoms. Methods: 43 patients with NPH and 129 community residents as control were recruited in this study and accepted a face-to-face questionnaire about risk factors, a clinical examination and magnetic resonance imaging (MRI) test during the visit.

Results:
The prevalence of stroke in NPH patients was significantly higher than that in control group. There were 6 patients with symptomatic normal pressure hydrocephalus (sNPH), whose causes were brain trauma, hemorrhage and meningitis. The vascular risk factors were similar in sNPH and iNPH group, in which the percentage of hypertension was significantly higher than control. The major first symptom of iNPH was gait disorder or cognitive impairment, which accounted for 78.38% and 21.62% respectively. The percentage of coexistence of three or two symptoms in iNPH were 45.95% and 40.54%, both of which were significantly higher than single symptom occurrence. The coexistence of gait disorder and cognitive impairment was 27.03%, although 10.81% patients had gait disorder and urinary incontinence simultaneously. Only 1 patient had cognitive impairment and urinary incontinence. Unexpectedly, there was no significant difference for sex or age between iNPH patients with one, two and three symptoms or with different first symptoms.

Conclusions:
Hypertension and history of stroke may be the major risk factors for NPH. There was no difference in cardiovascular risk factors between sNPH and iNPH. The results revealed different first and coexisted symptoms in iNPH patients, which may be parallel to each other.

O29
Identification of normal pressure hydrocephalus by disease-specific patterns of brain stiffness and damping ratio Introduction: Altered brain biomechanics, which can be measured noninvasively by magnetic resonance elastography (MRE), represent one hypothesis of normal pressure hydrocephalus (NPH) pathogenesis. Here we evaluated the accuracy of MRE-based viscoelasticity measurements to discriminate patients with NPH from both cognitively normal (CN) subjects and patients with probable Alzheimer's disease (AD). Methods: Thirty-three NPH, 44 CN and 8 AD subjects were scanned after obtaining IRB approval and written informed consent. MRE exams were collected with a spin-echo EPI pulse sequence (60-Hz motion, 3-mm isotropic final image resolution). Stiffness and damping ratio maps were computed using neural network inversion. A voxel-wise analysis was performed to compute maps of stiffness and damping ratio changes due to NPH. P < 0.025 was considered significant for each mechanical property (cluster-level family-wise error corrected). Each subject's MRE result was summarized by computing the correlation coefficient between that subject's age-and sex-corrected maps and the estimated group map while leaving out that subject. Results: Subjects with NPH exhibited a concentric pattern of stiffness changes with periventricular softening and stiffening near the dural surface. Damping ratio was also significantly decreased in NPH subjects. Introduction: Shunt failures are common and subject patients to multiple surgeries and decreased quality of life. A Shunt Outcomes Quality Improvement (ShOut-QI) initiative was implemented to reduce shunt failure incidence (SFI) through: (1) neuronavigationassisted proximal catheter insertion; and (2) laparoscopy-guided distal catheter anchoring over the liver dome to drain into the right upper quadrant (RUQ), away from omentum and common shunt obstruction-prone debris. A prospective cohort study tested the hypothesis neuronavigation and laparoscopy-guided VP shunt insertion (VPSI) will reduce incidence of shunt failure. Methods: "Pre-ShOut" and "Post-ShOut" groups of patients were assessed, which included those who had their initial VPSIs done before or after the ShOut-QI initiative, and without or with neuronavigation/laparoscopy, respectively. A 3-point CT index assessed proximal catheter placement, postop X-rays confirmed distal catheter placement, and a standardized protocol determined the primary outcome (SFI) as any return to surgery for shunt revision. Results: 244 patients (97 Pre-ShOut, 147 Post-ShOut), mean age 73 years, were observed for ~ 4 years. Neuronavigation improved proximal catheter placement accuracy by 20% (p < .001), and 90% of laparoscopy-guided distal catheters drained into the RUQ. SFI occurred in 57% vs 23%, with a mean duration of 380 vs 283 days to revision surgery, in the Pre-ShOut and Post-ShOut groups, respectively (p = .008). Conclusions: Adult SFI may be reduced by improving the accuracy of proximal catheter placement with neuronavigation and reducing the risk of distal catheter failure with neuronavigation-guided placement. Further studies are necessary to assess the effect of these interventions on long-term patient outcomes. Introduction: Gait disturbance is a major symptom of the idiopathic normal pressure hydrocephalus (iNPH), and it is assessed by many personnel including doctors and rehabilitation staff (rehabs). Here, agreement among multiple raters was examined using video-based gait analysis in patients with iNPH. Methods: Fifteen patients with definite iNPH were enrolled. Timed go and test (TUG) was done twice in all patients. The assessment of gait was done in 8 patterns including freezing, shuffling, wide-base and short-step. On the video-rating method, seven staff of 2 doctors and 5 rehabs assessed simultaneously. The iNPH grading scale (GS) was also scored. Agreement study was done with the Krippendorff alpha. The alpha value ≥ 0.67 was defined as good.

Results:
In the first assessment, no patterns were regarded as good in both all 7 staff and 4 staff (2 doctors and 2 rehabs), while agreement between 2 doctors and 2 rehabs, respectively, was good in some patterns. Good agreement was observed in the GS score both in all 7 staff and the 4 staff. After making consensus in gait patterns of iNPH, the second assessment was done. The consensus-making was helpful to improve agreement in some of patterns and GS in 4 staffs, but not in 7 staffs. Conclusions: Agreement study using Krippendorff alpha among multiple raters revealed that the agreement of gait patterns in the iNPH was not good with multiple raters, while the GS was useful in scale because of good agreement even for multiple raters.

Methods:
We have developed an on-line systematic questionnaire for follow-up of patients with hydrocephalus. The questionnaire is organized into domains (shunt-dysfunction symptoms, cognition, physical function, quality of life, etc.) with 3-6 validated indicator questions for each domain. Answers are flagged as "green" for no problems; "yellow" as possible need of clinical assessment or "red" as very likely/definite need of clinical assessment. All-green answers continue on-line followup; one or more "yellow answers" are interviewed by telephone and called in if necessary; one or more "red answers" result in an outpatient consultation.
Results: Screening of questionnaires has been successfully integrated into the work process in the hydrocephalus out-patient clinic. Unnecessary visits with absence from school/work/daily activities are avoided. Clinical follow-up has become systematically standardized.

Conclusions:
We believe that hydrocephalus follow up by on-line questionnaires is applicable in a much wider context. It may be particularly useful in cases of large geographical distances between home and clinic. Introduction: Recently, we have reported the usefulness of fluoroscopic-guided paramedian approach in LP shunt surgery. In this method, the catheter insertion success rate is 100%, and complications such as catheter insertion difficulty and nerve root pain are rare. This paper reports on practical tips for this method. Methods: LP shunt selection checklist: The main cause of nerve root pain was the passage of the subarachnoid catheter into the narrow spinal canal. Therefore, in cases with L2/3 or L1/2 spinal stenosis where the catheter passes, VP shunt surgery should be selected because of the high risk of nerve root pain. No nerve root pain has occurred since adopting this checklist. Skin marking method (2345 and 60 method): It is not easy for beginners to decide in which direction the puncture should be corrected even under fluoroscopy. This method facilitates the right correction of the puncture direction. Confirmation of the catheter in the fluoroscopic image: 100% prevention of caudal insertion of the catheter is possible. Valve placement to the optimal site and depth: Thick subcutaneous fat on the valve makes post-operative valve resetting and pumping reservoir difficult. To prevent these, the paramedian approach is performed from the contralateral side, and a valve is placed on the paraspinal muscle under a thin subcutaneous fat. One-pass passer method: This method prevents the valve from getting stuck during its insertion. Insert a peritoneal catheter and valve at the same time as making a subcutaneous pocket with a platelike shunt passer. stagnation of cerebrospinal fluid (CSF) turnover may cause amyloid-β peptide (Aβ) accumulation, which may be improved by shunt placement. Therefore, measuring high molecular weight Aβ42 oligomer (HMAβ) with at least nine subunits (≥ 30 kDa), could support differentiation of iNPH from Alzheimer's disease (AD), Parkinson's disease (PD), and progressive supranuclear palsy (PSP). It could also elucidate changes in amyloid aggregates in CSF after shunt placement. Methods: Fifty-three patients with NPH were included: healthy controls (HC, 30), AD (16), PD (14), and PSP (14). All patients with NPH had lumbo-peritoneal shunt (LPS); CSF samples were taken before and 1 year after surgery to measure phosphorylated tau (p-Tau), Aβ42, toxic Aβ42 conformer, and HMAβ via sandwich ELISA. NPH patients were divided into four subgroups: iNPH (18), NPH with AD pathology (17), NPH with Parkinson's spectrum (PS) (14), and NPH with AD pathology and PS (5) Introduction: Implantation of ventricular catheters for hydrocephalus shunts still remains a source of complications due to suboptimal or even wrong placement. The risk grows with smaller ventricles. Additional partially time-consuming tools like neuronavigation, Thomale guide, endoscopy etc. were used to improve the ventricular catheter placement. However, a simple, easy applicable tool is necessary for acceptance by everybody to improve the overall outcome. The solution might be mixed reality-an overlap of virtual reality with realityour very first experience is presented here.

Methods:
The new VSI-technology for HoloLens allows the surgeon to see the bore hole of the skull with the surface of the brain as well as the (holographic) superposed ventricles in the depth at the same time. With this visualization he is able to puncture the ventricle without the help of landmarks to two different planes of reference. Results: The first shunt surgeries with the augmented/mixed reality demonstrated the feasibility. The technique is convincing, although the method is new; only few minutes were needed for implementing the system.

Conclusions:
The technique of safe ventricular puncture is convincing and can be easily integrated in the surgical work-up to make shunt surgery safer. The potential is obvious in all surgical fields to guide needles, screws and other implants. The potential of reducing the overall complication rate of ventricle puncture of course has to be proven in randomized prospective trials. Introduction: Diagnosis of iNPH is based on clinical symptoms and structural changes in brain imaging including enlargement of lateral ventricles especially to the upwards direction. Widely used method for evaluation of that enlarging is determining the angle between the ventricles seen in coronal section (callosal angle, CA). Manual measurements are though laborious to use and suffer from subjective errors. That is why our goal was to create a fully automatic method for CA measurement. of 50 measurements were performed for each patient in that interval and the results were pooled to make a patient specific CAI-average. Normality of averages was tested, and t-test defined the statistical significance in differences between the groups. Results: CAI was statistically smaller in iNPH (average + standard deviation = 95.4 ± 12.9) compared with all the other groups including AD (121.8 ± 9.9, p = 1.7 * 10 −18 ), FTD (121.1 ± 9.0, p = 7.8 * 10 −11 ), LBD (118.2 ± 10.3, p = 0.019), MCI (119.5 ± 7.7, p = 2.1 * 10 −6 ), CN (120.7 ± 7.5, p = 1.4 * 10 −10 ), PD (113.8 ± 16.7, p = 0.042) and VaDS (119.2 ± 11.1, p = 2.3 * 10 −6 ). Conclusions: Automatic measurement of CAI provides potential fast and reliable method for differential diagnosis of iNPH. Introduction: In this retrospective study the idiopathic normal pressure hydrocephalus (iNPH) Radscale scores were assessed in brain computed tomography, with the purpose to evaluate the diagnostic accuracy of the iNPH Radscale. Methods: Seventy-five patients with iNPH, who had undergone ventriculoperitoneal shunt surgery and had been categorized as responders at clinical follow-up after 1 year, were compared with 55 asymptomatic controls (NPH score by Hellström > 90 points). One radiologist assessed the seven radiological features of the iNPH Radscale in computed tomography (CT) of the brain in the patients (preoperatively) and controls.

O41
Results: There was a significant difference between the shunted group and control group, with a mean iNPH Radscale score of 10 (IQR 9-11) and 1 (IQR 1-2) respectively, p < 0.001. Receiver operated characteristics analysis yielded an area under the curve of 99.7%, and a cut off level of iNPH Radscale score of 4 corresponded to a sensitivity of 100% and a specificity of 91%, with an overall accuracy of 96.2%.

Conclusions:
The iNPH Radscale can accurately separate shunt responsive iNPH patients from controls. This could be useful in excluding patients from the disease. Results: Compared with the negative group, the positive group tended to have higher Fazekas score of periventricular white matter (beta = 0.895, P = 0.068). In DTI analysis, the positive group had the significantly higher ADC value in the left ventricle posterior area lesions (P = 0.003), tended to have the higher FA value of the lesions in the right ventricle anterior area and the ADC value of the right ventricle posterior area lesions (P = 0.058, P = 0.058). The FA value of right ventricular anterior area was significantly correlated with motor function, cognitive and functional score and iNPHGS score. Conclusions: Periventricular white matter lesions in patients with idiopathic normal pressure hydrocephalus are significantly correlated with their clinical features, which may be one of the pathogenesis mechanisms and the target for improving symptoms after drainage surgery.

Introduction:
We investigated the possibility of using protein tyrosine phosphatase receptor type Q (PTPRQ) for auxiliary diagnosis of idiopathic normal pressure hydrocephalus (iNPH), and carried out the first intracerebral analysis of PTPRQ expression in autopsied brains of patients with iNPH.

Methods:
We analyzed the feasibility of using PTPRQ concentrations in the cerebrospinal fluid (CSF) for auxiliary diagnosis of iNPH in the Finnish (n = 30) and Japanese (n = 30) population. PTPRQ concentrations in iNPH patients and healthy elderly subjects with normal cognition (NC, n = 40) were compared. PTPRQ expression levels were measured in autopsied brains of iNPH patients and NC subjects. Results: PTPRQ concentration was increased by iNPH; the concentration was higher in the Finnish-iNPH (mean 762 [SD 570] pg/mL) and the Japanese-iNPH (712 [832]) groups than in the NC group (351 [100]) (p < 0.001 and p = 0.018, respectively). In a combined Finnish and Japanese iNPH group, using a PTPRQ cutoff of 370 pg/mL, iNPH was detected with a sensitivity, specificity, and area under receiver operating characteristic curve of 75%, 65%, and 0.771, respectively. Conclusions: Measurement of PTPRQ in the CSF by ELISA showed levels approximately 2 times higher in patients with iNPH than in healthy elderly subjects, regardless of the racial group, confirming the validity of this assay for auxiliary diagnosis. The absence of a relationship between PTPRQ and p-Tau, t-Tau, and Aβ42 markers of AD pathology, and the high levels of PTPRQ in patients with iNPH have an important diagnostic merit. Results: In the control group, the peak radio activities of GM, WM and BG were at 22.5, 50.0 and 22.5 s after the peak in ICA, respectively. Activities in whole brain structures decreased gradually. On the contrary, activities of LV, FS and PPC increased gradually until the end of measurement. In the iNPH group, RRA of BG was significantly lower than controls. RRA curves of GM and WM were decreasing and also getting closer, each other in late phase. This means diffusion of water molecules in brain resulting in equal distribution in time. Compared with the controls, it took significantly longer until the equal distribution in brain in iNPH. RRA of LV, FS and PPC in iNPH tended to be lower compared to controls. After L-P shunt, these delays tended to be normalize. Conclusions: Water turnover in brain and CSF is reduced or delayed in iNPH compared to normal, which is normalized after shunt surgery. Introduction: Much has been published and attempted to improve EVD survival and clearance of haemorrhagic CSF from the ventricles in intraventricular haemorrhage. They do represent a very particular type of blockage, causing an increased morbidity. We applied and followed two different methodologies of external ventricular drainage by using two different concepts of EVD catheters and compared the effectiveness of each method in relation to its impact in the number of hospitalization days, intensive care days (I.C.U.), hydrocephalus occurrence and patient's general outcome. Methods: Electronic files of 30 patients with intraventricular haemorrhage between March 01, 2014 and April 30 2015 were analysed. They were divided in two groups: group H (15 patients who used catheters LCR600H and group C (15 patients who used conventional catheter LCR600A). The only difference was the design of the catheter.

Results:
There was a significant difference on number of days that patients remained at the (I.C.U.) in favour of group H compared to group C (p < 0.01). Also, group H ventricular catheters were withdrawn earlier than group C catheters (p < 0.01). Conclusions: Results allows us to conclude that the LCR600H catheter was able to show some advantages for influencing the risks exposure and permanence in the ICU compared to conventional catheters. Current results warrants a more detailed and multicentric study to evaluated social and financial impact on the Brazilian Health care.
Introduction: Brain compliance in hydrocephalus remains a controversial issue; i.e. stiffness reportedly increases in NPH and decreased stiffness may contribute to ETV failure. We used Magnetic Resonance Elastography (MRE), a new tool to measure brain stiffness non-invasively, to determine if stiffness is altered in pediatric hydrocephalus.
Methods: From 2 centers, 40 shunt-dependent patients (age 0.6-39 year, median 18.0) who developed hydrocephalus as infants were compared to 27 healthy age-matched controls (age 6-46 year, median 16.7). MRE was performed by inducing a 30 Hz vibration transmitted through the skull. Tissue elastance (G*, inverse of stiffness) averaged separately across white and grey matter masks and within lobar regions was calculated through Algebraic Helmholtz Inversion. The Headache Disability Index (HDI) and Hydrocephalus Outcome Questionnaire (HOQ) were collected in all Einstein patients. Results: In periventricular white matter, brain tissue stiffness was reduced significantly (p < 0.005) in patients compared to controls (G* = 1.75 ± 0.28 kPa vs. 1.97 ± 0.22 kPa). Occipital grey matter stiffness correlated negatively with ventricular size (R 2 = 0.23, p < 0.001).
There was a weak positive correlation between occipital grey matter elastance and HOQ (R 2 = 0.16, p < 0.05), and a negative trend correlation between occipital grey matter stiffness and HDI (R 2 = 0.14, p = 0.056). One patient scanned 1 day prior to shunt revision, and 11 months following revision, exhibited increased stiffness (i.e. toward controls) in all lobes except the occipital lobe where stiffness decreased. Conclusions: Brain stiffness was reduced in hydrocephalus patients, suggesting impaired biomechanical integrity of brain tissue. Introduction: In Japan, iNPH patients prefer lumbo-peritoneal (LP) shunt more than ventriculo-peritoneal shunt recently. LP shunt does not need brain surgery and is less invasive. However, LP shunt has some pitfalls. Post-operative abdominal catheter migration was reported in Sinphoni 2 study and then abdominal catheter has been sutured to the abdominal fascia in our institute. We experienced another complication of lumber catheter migration during surgery. Intraoperative CSF out-flow was good, and we didn't notice the epidural migration during the surgery. So, we started intraoperative shuntography. Methods: From April 2015 to September 2017, 32 iNPH cases were lumbo-peritoneally shunted with intraoperative shuntography. After test tap with 21-gauge lumbar needle, Tuohy needle was inserted.

O50 Do co-morbidities influence shunt outcomes in idiopathic intracranial hypertension?
During the lumbar catheter insertion through the Tuohy needle, 1 to 3 ml of the water-soluble nonionic iodinated contrast agent was slowly injected to visualize the tube form and surrounding structure. When the lumber catheter was in wave or hair pin form, or in the epidural cavity, tube insertion was re-performed. Also, when the abdominal catheter was suspected to be malpositioned in one case, abdominal shuntography was also performed. Results: In 32 cases, there was no complication associated with shuntography. During the observation period, one cerebellar hemorrhage, one cerebral infarction, one prolonged intracranial hypotension which lasted over a week and one catheter disruption was found.
The lumbar catheter was in the epidural cavity in two cases and the abdominal catheter was outside the peritoneal cavity in one case. Those were successfully replaced. Conclusions: Intraoperative shuntography is safe and it reduced catheter migration in lumbo-peritoneal shunts.
Introduction: Postural instability, often with a forward leaning posture, is a common feature in idiopathic normal pressure hydrocephalus (iNPH). However, the spinopelvic alignment and presence of concomitant spinal sagittal plane deformity has not been well studied in the iNPH population. The objective of this study was to measure the baseline spinopelvic parameters and determine the prevalence of sagittal plane spinal deformity in patients with iNPH. Methods: We reviewed a series of patients who underwent VP shunting for the treatment of iNPH and who also had standing scoliosis x-rays. We evaluated for comorbid spinal deformity based on the SRS-Schwab adult spinal deformity classification system by assessing Fluids Barriers CNS 2019, 16(Suppl 3):36 pelvic incidence minus lumbar lordosis (PI-LL), pelvic tilt (PT), and sagittal vertical axis (SVA). Results: Seventeen patients with iNPH were included for analysis. Six patients (35%) met criteria for having marked spinal deformity by at least one radiographic parameter: 5 (29%) had greater than 20° PI-LL mismatch, 3 (18%) had > 9.5 cm SVA, and 1 (6%) had PT exceeding 30°. Additionally, the degree of thoracic kyphosis exceeded that of lumbar lordosis in 9 patients (53%). Conclusions: Sagittal plane spinal deformities may be common in iNPH patients and may contribute to postural instability in this patient population. Moreover, since thoracic kyphosis exceeded lumbar lordosis in more than half of the patients, the forward leaning postural instability in iNPH patients may be due to increased thoracic kyphosis. These findings warrant further investigation along with a determination if there is any change in the parameters following shunt placement. Fluids Barriers CNS 2019, 16(Suppl 3):36 Introduction: CSF is thought to help large molecules to move in and out of the brain, but exactly how it is done is largely unknown.
We tried to acquire some information about large molecule transport by CSF, by closely analysing the T2 of the CSF. We assume that the T2 of the CSF varies with protein content, the higher the content, the shorter the T2. Methods: Multiple spin-echo images were acquired using a CPMG (Curr, Purcell, Meiboom and Gill) imaging sequence. Since each voxel may contain many T2 components, due to either intra-voxel small structures or due to simple partial volume effect, the decaying signal was decomposed into many components using NNLS (nonnegative least squares) decomposition. Results: 25 echoes were acquired with echo interval of 40 ms. The resulting echo times for each image were: 40, 80,…, 1000 ms. The decaying signal was decomposed into 25 components, with pixel by pixel basis. Components with 300 ms or longer T2 were considered to be CSF, without partial volume effect. The CSF component had mostly a single T2 group, and average T2 was calculated for each pixel. Even within the components which purely consists of CSF had a wide range of T2 values, from 500 to 2000 ms, and the spatial variation of the average T2 values was visualised, which is considered to correlate to protein content of the CSF. Conclusions: A method to estimate T2 component of CSF without partial volume effect was developed. Introduction: Vitamin D has multiple functions in the central nervous system. Especially, many studies reported that decreased serum 25-hydroxyvitamin D (25OHD) concentrations may be associated with cognitive disorders and larger lateral cerebral ventricles. However, there has been no report about the relationship between the vitamin D concentration and idiopathic normal pressure hydrocephalus (iNPH). The purpose of this study was to investigate the effect of vitamin D concentration on the clinical prognosis in iNPH after surgery.

O63 Clinical significance of vitamin D concentration in idiopathic normal pressure hydrocephalus after shunt surgery
Methods: This research was conducted by Korea Brain Bank Network Project operated through Korea Brain Research Institute funded by the Ministry of Science and ICT. Between 2016 and 2018, 37 patients with iNPH underwent ventriculoperitoneal shunt surgery. Serum 25OHD concentration was quantified at shunt surgery. The patients were categorized into two groups, such as low 25OHD group (< 15 ng/ml) and high 25OHD group (> 15 ng/ml), and analyzed in terms of clinical and radiological findings. Results: Thirty-seven patients consisted of 19 patients with low 25OHD concentration (mean, 73.1 ± 6.4 years; 36.8% female) and 18 patients with high 25OHD concentration (mean, 74.7 ± 4.3 years, 44.4% female). The relationship between the 25OHD concentration and Mini-Mental Status Examination showed the positive correlation with a statistically marginal significance (r = 0.299, p = 0.06). Additionally, 5 chronic subdural hematomas (CSDH) after shunt surgery were developed, and all of them with CSDH were included in low 25OHD group (p = 0.046). Conclusions: Low serum 25OHD concentrations may be associated with cognition impairment before shunt surgery and CSDH after shunt surgery in iNPH.

O64
Performance analysis of the initial pressure selection protocol for the sphera pro programmable valve Fernando C. G. Pinto, Rodolfo C. Reis, Manoel J. Introduction: The correct choice of initial valve programming at the time of shunt implantation improves the clinical outcome. The objective of this study was to evaluate the efficiency of the initial adjustment protocol for Sphera Pro valve. Methods: Twenty-four patients underwent surgical treatment for implantation of the Sphera Pro valve. The protocol for initial choice of valve programming was elaborated considering the diagnosis of the hydrodynamic disorder: NPH and arachnoid cyst at 3 cmH2O, pseudotumor cerebri at 16.5 cmH2O. The antigravitational device implanted in all patients was 15 cm H2O, except for arachnoid cyst. Results: Twenty NPH patients were included in the study and presented with progressive clinical improvement, with the mean scores on the Japanese scale being 5.7, 3.9, 2.6, 1.3, 1.3 corresponding to the pre, 10 days, 3, 6 and 12 months postoperative periods. Three patients presented with improvement after reduction of the pressure from 3 to 1 cm H2O after 3 months, one patient presented with overdrainage with formation of subdural collection and reversed with pressure adjustment to 21 cm H2O for 30 days and progressive opening up to 10 cm H2O. Three patients with pseudotumor cerebri were included in the study and improved, one needing to have the valve removed by infection. A patient with arachnoid cyst presented clinical improvement and did not require adjustment.

Conclusions:
The protocol was efficient in 83.3% of the cases. The valve pressure initially chosen was maintained in 20 of the 24 patients for a period of 1 year, with clinical evidence of progressive improvement. Introduction: Ehlers-Danlos syndrome (EDS) is known to be associated with cerebrospinal fluid (CSF) disturbances, including recurrent CSF leak and Chiari-I malformations. Persistent pseudomenginocoeles are known to be associated with raised ICP. We present an unusual case of a compressive epidural CSF collection after CT-guided L5 nerve root block and describe the effective management strategy. Methods: Retrospective case report. ICP was monitored using Liquoguard ® 7 (Moller Medical GmbH). ICP data was processed using Excel (Microsoft) and on GraphPad Prism 6.0c.

Results:
A 29 year old female known for EDS presented with a symptomatic L4-S2 epidural compressive CSF collection 5 weeks after CTguided L5 nerve root block. Eleven days after decompression and repair, she developed a symptomatic subcutaneous pseudomeningocoele, for which she had a second repair and insertion of a lumbar drain, that was then connected to a Liquoguard ® 7 (Moller Medical), a pressure and volume controlled CSF drain that enables ICP monitoring with a safety mechanism to prevent over-drainage, due to her Chiari-I malformation.
Median ICP decreased from 24 mmHg after post-repair to 12 mmHg on day 7, before drain removal. There was no further CSF leak or collection at 3 months. Conclusions: EDS in the presence of raised ICP presents a unique challenge as the propensity for CSF leaks due to frail dura is exacerbated by the raised pressure driving the CSF leak. Digitally pressure and volume controlled lumbar drainage can assist with decisions regarding Fluids Barriers CNS 2019, 16(Suppl 3):36 detailed spatial-temporal distribution of solute concentration over 24 h. Methods: CSF geometry was reconstructed from T2-weighted MRI of a healthy 23-year-old female and included realistic spinal cord nerve roots as well as key intracranial CSF spaces. 3D-printing was used to construct the complete geometry in a transparent material. Timelapse imaging was used to capture spatial-temporal distribution of fluorescein. Distribution was analyzed under different injection conditions (e.g. injection volume and location, filtration loop, and others) with various CSF flow waveforms (e.g. magnitudes, frequencies). A computational fluid dynamics (CFD) study of the pulsatile in vitro CSF flow field was solved using ANSYS Fluent by computing the steadystreaming velocity field. In vitro results for a specific injection scenario were compared to the CFD simulations by linear regression of the average tracer concentration for 3 mm thick axial slices. Results: Total spinal and intracranial CSF volumes were 100.3 ml and 221.6 ml respectively. Maximum Reynolds number was 461. CFD predicted steady streaming velocities in the cranial subarachnoid space were ~ 50X smaller than in the spine. Agreement of in vitro versus CFD spatial-temporal solute concentration was strong for all injection scenarios analyzed.
Introduction: Approximately 37% of long-duration spaceflight astronauts develop signs/symptoms of the spaceflight associated neuroocular syndrome (SANS), including optic disc edema, chorioretinal folds, ocular globe flattening and hyperopic shifts. Quantification of ophthalmic changes that occur during spaceflight may provide clues into the mechanisms responsible for SANS. Automated and manual methods were developed to quantify optic nerve (ON), optic nerve sheath (ONS), and optic globe geometry to better understand how microgravity may impact these structures. Methods: Magnetic resonance (MR) images were collected from astronauts before and after long-duration spaceflight. 3D ON and ONS geometries were analyzed using threshold-based segmentation to compute cross-sectional area. Threshold segmentation was applied to the optic globe after radially re-slicing MRI sequences.
Resulting pre-and post-flight point clouds were aligned using an iterative closest point algorithm. Posterior ocular globe flattening was assessed in terms of volume deformation at a radius of 4 mm around the ONH. Results: No significant changes were observed in ON and ONS geometries after long-duration spaceflight, however some astronauts did exhibit significant flattening of the posterior ocular globe. The average and standard deviation of the posterior globe volume deformation was − 8.3 ± 9.1 mm 3 (p = 0.0001, N = 20 eyes). Notably, the subject with the greatest degree of posterior ocular globe volume deformation (39.2 mm 3 ) was clinically diagnosed grade 1 optic disc edema via fundus imaging. The role of intracranial pressure changes in astronauts presenting with ocular globe deformation in astronauts is unknown. Introduction: The issue of cortical atrophy is important in normal aging and disease since it is associated with cognitive and physical impairments. Cortical atrophy is potentially a relevant biomarker for the early diagnosis of Alzheimer's disease (AD). The vascular component is also an integral part of AD and other late-life neurodegenerative diseases. Abnormalities in blood flow appear before accumulation of abnormal proteins in AD. The occlusion of capillaries by neutrophils is significantly higher in AD animal models than control and reduction of those occlusions with an antibody increases both blood flow and cognitive capacities. Vascular alterations lead to hypoperfusion, oxidative stress and inflammation, which in turn lead to damage of neurons, glia and myelin, predominantly in the white mater. Implication of vascular pathologies for gray matter remains unclear. A recent study showed that altered cerebral hemodyamics in asymptomatic carotid artery stenosis is associated with cortical thinning. However there is no proven link between vascular pathologies and cortical thinning. We propose to explore brain aging with a combined biomechanical and imaging approach in order to assess both fluid dynamics alterations and brain structural modifications. We hypothesize that there is a link between altered cerebral hemodynamics and loss of cortical thickness during brain aging. Methods: 80 patients suspected of hydrocephalus were prospectively involved. All patients complain of gait alteration, urinary difficulties, mild apathy and ventriculomegaly on brain imaging. They underwent brain MRI with T1 weighted images to quantify cortical thickness and phase contrast images to measure arterial, venous and CSF velocities. Lumbar infusion test was also performed to gauge lumbar pressure, a surrogate marker of intracranial pressure (ICP), and CSF dynamics. The cortical volumetric segmentation was done by an automatic post-processing analysis with FREESURFER. Venous, arterial and CSF velocities were measured from PCMRI with BIOFLOWIMAGE software. ICP and CSF dynamics were extracted from infusion tests. Pearson correlations were calculated between cortical thickness and arterial, venous and CSF velocities, but also ICP and derived indices. Results: Mean cortical thickness was positively correlated with mean ICP (r = 0.48, p = 0.001), ICP pulse amplitude (r = 0.43, p = 0.001), arterial flow (r = 0.44, p = 0.001), aqueductal CSF flow(r = 046, p = 0.001), but negatively correlate with venous flow (r = -0.44, p = 0.001) Conclusions: We demonstrate that cortical thickness is correlated with arterial and CSF pulsatility. The causality is more complex; however the association between intracranial pulsatility and gray matter thickness suggests that there is a relationship between vascular alterations at the macroscale level and the pathobiology of cortical atrophy. Introduction: Chronic post-traumatic headache (CPHT) and/or mild traumatic brain injury (mTBI) are serious sequela of head injury (JAMA 300;711-9, 2008, J Neurotrauma 34;1524-30, 2017. Post-traumatic headache in whiplash-associated disorder was suggested to have orthostatic nature similar to that of spontaneous intracranial hypotension (Anesth Analg 105;809-14, 2007). We investigated the intracranial pressure (ICP) in CPTH and studied the effect of cerebrospinal fluid (CSF) removal in these patients. Methods: ICP was measured by lumbar puncture in consecutive 279 CPTH patients. CSF was removed unless the patient did not complaint headache. Data were shown in mean (SD), Student's t-test was used and the statistic significant level was set p < 0.03. Methods: A strategy employing voxel-based classification through AI and features derived from positional and intensity constraints was used to determine the lateral and third ventricular volume from MR images. Using the same MR imaging data, 3D printed models were created whose volumes were determined using a precise water displacement technique and compared with the volume as measured by AI. The subjects were five pediatric patients with normal size ventricles and three with hydrocephalus.

Results:
The correlation between the AI determined ventricular volume and those of the 3D models was between 87 and 94% (Jaccard index of similarity).

Conclusions:
The accuracy of ventricular volume determination is expected to improve with further AI machine learning. When implemented, accurate ventricular volume can be automatically be made a part of the radiology report. Volume and geometrical features of anything that can be segmented should be able to be quantified. Summary: We previously reported a novel heritable form of hydrocephalus termed ETINPH (essential tremor-idiopathic normal pressure hydrocephalus) in a large 5-generation pedigree, mapped this ETINPH locus to chromosome 19q12-13.31, narrowed down ETINPH gene to 25 genes on chr 19 with WES (Whole Exome Sequencing), and eventually further narrowed down to 8 genes with non-synonymous mutations. In this report, we employed NGS (Next-generation sequencing) technologies to define the genomic structure of a new ETINPH patient from an unrelated ETINPH pedigree identified recently. NGS was performed with approximate 100X coverage and 100 bp paired-end reads in Illumina 2000. All reads were mapped to human genome GRCh38 to generate genetic variants files, SNP and INDEL. Newly identified genetic variants (SNPs and INDELs) have been compared with the known 8 candidate mutations in the original family with functional prediction, pathway analysis and interactome. The molecular function of ETINPH and its associated pathway will be discussed. Introduction: Vertebrobasilar dolichoectasia (VBD) is a condition encountered in the elderly population characterized by marked elongation, dilatation, and tortuosity of the vertebral and basilar arteries. Arteriosclerosis and hypertension are risk factors that are believed to be partly involved in the pathogenesis. Obstructive hydrocephalus has been described as an infrequent condition of VBD, usually caused by compression of the third ventricle. We report two patients with Normal Pressure Hydrocephalus (NPH) and VBD but without an obstruction raising questions about a possible link between these conditions. Methods: Case 1 was a 72-year-old male and the case 2 was a 71-yearold man. Both of them were referred to our outpatient clinic with cognitive impairment, gait disturbance and urinary urgency. Both patients underwent radiological brain investigations with Computer Tomography (CT), CT-angiography and Magnetic Resonance Imaging which were compatible with NPH and VBD, without appearance of any compression of third ventricular outflow. In addition, there were also periventricular microangiopathic changes in the patients. Both of them fulfilled the international criteria for idiopathic normal pressure hydrocephalus and were improved after shunt surgery. Results: VBD is an arterial disease, mostly attributed to arteriosclerosis. In literature, cases of VBD caused by an obstructive hydrocephalus have been described. However, there is some evidence of a disturbed CSF-dynamics in hydrocephalic patients with VBD but, without an obstruction. Those two patients with NPH and VBD raise the question about a potential link and possible overlooked cause of NPH. Further studies, especially on CSF-dynamics, are needed to clarify this.

Introduction:
When considering the underlying pathophysiological mechanisms involved in idiopathic normal pressure hydrocephalus (INPH), white matter is often the main locus of investigation. However, when an axon in the brain is damaged, degeneration of the neuron can occur proximally (dying back) and Alzheimer's disease, associated with cortical thinning, is a common pathologic comorbidity with INPH. We investigated differences in cortical thickness between healthy controls and INPH patients who had a positive response to the CSF tap test. We also evaluated relationships between cortical thinning and Gait Status Scale in INPH patients. Methods: Forty-nine patients diagnosed with INPH and 26 healthy controls were imaged with MRI, including 3-dimensional volumetric images for cortical thickness analysis across the entire brain. Results: INPH patients, when compared to control subjects, showed statistically significant cortical thinning in the left superior frontal gyrus (orbital part), left superior frontal gyrus (medial orbital part), bilateral gyrus rectus, right insula, bilateral parahippocampal gyrus, left fusiform gyrus, right heschl gyrus, right superior temporal gyrus, bilateral temporal pole of the superior temporal gyrus, bilateral middle temporal gyrus, bilateral temporal pole of the middle temporal gyrus, and bilateral inferior temporal gyrus after FDR correction (p < 0.05). Cortical thinning of the right superior frontal gyrus (medial orbital part), right gyrus rectus, right insula, right temporal pole of the superior temporal gyrus, and right superior temporal gyrus was correlated with worse performance in the Gait Status Scale (p < 0.01 uncorrected).

Conclusions:
Our results indicate that INPH might be a disease exhibiting a characteristic pattern of cortical thinning.

P3
Lumbo-peritoneal shunt surgery with initial valve setting "virtual off mode" for iNPH patient Takashi Kawahara 1 , Masamichi Atsuchi 1 , Takuichiro Higashi 2 , Ryuji Awa 2 , Takuya Iwasaki 1 , Koji Yoshimoto 2 1 Division of Neurosurgery, Atsuchi Neurosurgical Hospital, Kagoshima, Japan; 2 Department of Neurosurgery, Graduate School of Medical and Dental Sciences, Kagoshima University, Japan Correspondence: Takashi Kawahara -takashi.kawahara@jifukai.jp Fluids and Barriers of the CNS 2019,16(Suppl 3):P3 Introduction: Over drainage after lumboperitoneal shunt (LPS) surgery might cause intracranial hypotension. Sometimes, it would induce severe subdural hematoma. To prevent of this complication, controllable valves are available. However, some patients suffer these complications in the condition of highest valve setting. We report our experience of "Virtual off mode" of the Codman CER-TAS Plus Valve for initial valve setting in the LPS surgery for idiopathic normal pressure hydrocephalus (iNPH) patients. We describe the usefulness of highest valve setting of controllable valve for initial valve setting of LPS. Methods: A single-center retrospective study of iNPH patients undergoing LPS procedure with initial valve setting 8 of CERTAS Plus valve Fluids Barriers CNS 2019, 16(Suppl 3):36 from December 2018 to April 2019. Patients' records were retrospectively reviewed for clinical and subjective outcomes. Results: Continuous 21 iNPH patients underwent LPS surgery with initial valve setting 8 of CERTAS Plus valve. A month after LPS surgery, nineteen cases (90.5%) presented good outcome for NPH symptoms. Seven patients (33%) were not necessarily set lower and setting was kept at 8. There were no severe complications after LPS surgery. For example, postural headache, subdural effusion, chronic subdural hematoma. We compared the outcomes of initial setting of programmable valve seven, we already reported at Hydrocephalus 2019 and the outcomes of setting eight (Virtual off mode). As Virtual off is not completely off, the improvement rate of the symptoms was the same as initial setting 7. And the rate of over drainage is less than initial setting 7. So we think that Virtual off mode is recommended as the initial setting of LPS for iNPH patients. Conclusions: Virtual off mode of Codman CERTAS Plus valve is available for initial valve setting for LPS surgery of iNPH patients. This strategy avoid severe over drainage symptoms and severe subdural hematoma. But some cases were not completely improved, as virtual off setting. So we recommend to adjust the valve setting after 1 week after surgery. Introduction: The aim of the study was to detect the preoperative factors of poor outcomes after shunt surgery in patients with idiopathic normal pressure hydrocephalus (iNPH). Methods: Eighty-eight consecutive patients with iNPH who underwent shunt surgery were enrolled in this study. Shunt responsiveness was defined as an improvement by one or more points on modified Rankin Scale at 1-year follow-up after surgery. To evaluate patients' symptoms, we administered the Mini-Mental State Examination (MMSE), Frontal Assessment Battery (FAB), the timed up and go test (TUG), and the iNPH grading scale (iNPHGS) before and 1 year after surgery. We also assessed the presence of risk factors for cerebrovascular diseases, including hypertension, diabetes mellitus, hyperlipidemia, and smoking. Results: Thirty-eight patients (43.2%) had good outcomes to surgery (responders), and 50 patients (56.8%) had poor outcomes (nonresponders). The prevalence of hypertension in the non-responders was significantly higher than that in the responders. In addition, the median score of iNPHGS urinary dysfunction in the non-responders was significantly higher than that in the responders. Moreover, the logistic regression analysis revealed that the best predictor of poor outcomes was the presence of hypertension (odds ratio = 3.324, 95% CI = 1.334-8.282). Conclusions: There is a possibility that irreversible brain damage in iNPH is facilitated by hypertension. In addition, severe urinary dysfunction in iNPH might be a sign of irreversibility of the brain damage and/ or be related to co-morbidity with other neurodegenerative disorders that cause urinary dysfunction. Fluids Barriers CNS 2019, 16(Suppl 3):36 • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress.

Learn more biomedcentral.com/submissions
Ready to submit your research ? Choose BMC and benefit from:

Conclusions:
The correlation mapping technique in conjunction with S-transform and real-time 2D-PC imaging was useful for characterizing CSF dynamics in the intracranial space. Introduction: Modelling brain changes in disorders of cerebrospinal fluid flow requires segmentation of the tissue compartments that abnormal flow tends to disrupt. The morphological abnormalities frequently present in such disorders limit the accuracy and robustness of traditional computational methods. The difficulty arises from the extreme heterogeneity of possible anatomical appearances, which a standard template-the reference for conventional methods-is likely to find great difficulty with. Here we sought to exploit the greater expressive power of deep learning-based methods to create a fully automated algorithm for tissue segmentation that is robust to a wide range of pathological changes, evaluating it on real-world hospital magnetic resonance imaging. Methods: We investigated an unselected set of 301 consecutive patients with cerebrospinal fluid (CSF) flow disorders who had undergone routine magnetic resonance imaging and intracranial pressure monitoring at the National Hospital for Neurology and Neurosurgery. A 3D convolutional artificial neural network architecture was trained and optimized on brains from OASIS-3, and a subset of the clinical scans, to generate probabilistic tissue partitions of grey & white matter, and CSF. Tissue segmentation based on SPM12 was used as a reference conventional approach. Results: We quantified the differences in performance of the two approaches across the dataset, categorized by diagnosis, by visual inspection and comparison with manually defined tissue compartments. Conclusions: Deep learning methods of tissue segmentation are promising, robust alternatives to the traditionally used methods, such as SPM software, which can be brittle when asked to segment anatomically deformed brains.

Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.