Application of MR virtual endoscopy in children with hydrocephalus

    Application of MR Virtual Endoscopy in Children with Hydrocephalus Cailei Zhao, Jian Yang, Yungen Gan, Jiangang Liu, Zhen Tan, Guohua Liang, Xianlei Meng, Longwei Sun, Weiguo Cao PII: DOI: Reference:

S0730-725X(15)00189-7 doi: 10.1016/j.mri.2015.07.013 MRI 8400

To appear in:

Magnetic Resonance Imaging

Received date: Accepted date:

10 February 2015 29 July 2015

Please cite this article as: Zhao Cailei, Yang Jian, Gan Yungen, Liu Jiangang, Tan Zhen, Liang Guohua, Meng Xianlei, Sun Longwei, Cao Weiguo, Application of MR Virtual Endoscopy in Children with Hydrocephalus, Magnetic Resonance Imaging (2015), doi: 10.1016/j.mri.2015.07.013

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Application of MR Virtual Endoscopy in Children with Hydrocephalus

RI P

T

Running title: a non-invasive diagnostic approach

SC

Cailei Zhaoa, MM, Jian Yangb*, MD, Yungen Gana, BM, Jiangang Liua, MD, Zhen

Caoa, MM

a

MA NU

Tana, MM, Guohua Lianga, BM, Xianlei Menga, BM, Longwei Suna, MM, Weiguo

Department of Radiology, Shenzhen children’s Hospital, No. 7019, Yitian Road,

Department of Radiology, The first affiliated hospital of Xi’an jiaotong university,

PT

b

ED

Shenzhen, 518038, China.

CE

No. 277, Yantaxi Road, Xi’an, 710061, China.

AC

*Corresponding to: Jian Yang, Department of Radiology, The first affiliated hospital of Xi’an Jiaotong University, No. 277, Yantaxi Road, Xi’an, 710061, China. Tel: 86-18991232396, Fax: 029-85323643. E-mail: [email protected].

1

ACCEPTED MANUSCRIPT Abstract

RI P

T

Purpose: To evaluate the performance of MR virtual endoscopy (MRVE) in children

SC

with hydrocephalus.

Materials and Methods: Clinical and imaging data were collected from 15 pediatric

MA NU

patients with hydrocephalus and 15 normal control children. All hydrocephalus patients were confirmed by ventriculoscopy or CT imaging. The cranial 3D-T1 weighted imaging data from fast spoiled gradient echo scan (FSPGR) were transported to

ED

working station. VE images of cerebral ventricular cavity were constructed with

PT

Navigator software.

CE

Results: Cerebral ventricular MRVE can achieve similar results as ventriculoscopy in

AC

demonstrating the morphology of ventricular wall or intracavity lesion. In addition, MRVE can observe the lesion from distal end of obstruction, as well as other areas that are inaccessible to ventriculoscopy. MRVE can also reveal the pathological change of ventricular inner wall surface, and help determine patency of the cerebral aqueduct and fourth ventricle outlet.

Conclusion: MR virtual endoscopy provides a non-invasive diagnostic modality that can be used as a supplemental approach to ventriculoscopy. However, its sensitivity and specificity need to be determined in the large study. 2

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

RI P

T

Keywords: Approach, Application, Children, Hydrocephalus, MR, Virtual Endoscopy

3

ACCEPTED MANUSCRIPT 1. Introduction

RI P

T

Hydrocephalus is a common disease in children. Its diagnosis can be easily made based on the morphological changes of ventricular expansion. However, determining

SC

its etiology, type and degree of obstruction, and early diagnosis prior to morphological change remain challenging. Clinically accurate diagnosis and early treatment of

MA NU

obstructive hydrocephalus is critical for its prognosis [1, 2]. Current MR technologies, such as MR conventional scan, enhanced scan, and 3-Dimensional constructive inference in steady state (3D-CISS), Phase Contrast cine (PC-cine), Turbo-Spin Echo

ED

(TSE) and Gradient Recalled Echo (GRE) T2 sequences, mainly diagnose and guide

PT

the treatment of hydrocephalus on the basis of ventricle structural morphology [2-4]. Of them, 3D-CISS is widely used and sensitive to obstructive membranes. But when

CE

there is no significant obstructive membranes in the ventricle, its sensitivity is reduced

AC

and it becomes difficult to observe subtle abnormal changes of ventricle wall [2, 4]. PC-cine can be used to measure cerebrospinal fluid (CSF) dynamics and determine the type of obstruction in hydrocephalus. However, because in patients of different ages the cerebral aqueduct flow or direction varies, and many factors including the coding of velocity can affect the accuracy of flow velocity measurement[5, 6], early and accurate diagnosis of hydrocephalus remains to be difficult. Ventriculography is occasionally performed to determine the type of obstruction in hydrocephalus patients. But as an invasive procedure, it has inherited risk of severe complication, false positive result and is yet to be widely accepted [7]. This study explores a non-invasive 4

ACCEPTED MANUSCRIPT MR technique for the diagnosis of hydrocephalus.

RI P

T

MRVE is a virtual computer imaging technology used to visualize lumina inner surface. It collects data from continuous thin-slice MR scans and transmits them to the

SC

backstage processing station, where the computer software process the data to construct three-dimensional images of lumina inner surface, similar to imagines of

MA NU

fiberoptic endoscopy. MRVE reflects the object’s three-dimensional surface morphology. This technique has been successfully applied in the examination of hollow viscus (airway, gastrointestinal tract, bladder, etc.) [8, 9] as well as blood

ED

vessel, colon, delicate structure of internal auditory canal and the joint cavity [10-13].

PT

However, its use in observing the morphological and pathological changes of ventricular inner wall surface has yet to be reported. This study evaluates the

CE

performance of MRVE in visualizing pathological changes of the inner surface of

AC

cerebral ventricles and determining the diameter of the lumen in hydrocephalus patients. It evaluates the performance of MRVE in children with hydrocephalus by comparing its

findings

with

the results

of CT ventriculography and/or

ventriculoscopy.

2. Materials and Methods

2.1 Study Population We collected clinical and imaging data from 30 subjects consecutively enrolled from 5

ACCEPTED MANUSCRIPT May 2012 to December 2012. Of them, 15 were hydrocephalus patients, including 7 males, age range from 1 to 20 months with an average of 7.7 months; 8 females with

RI P

T

an average age of 7.6 months. All had clinical manifestation of enlarged head circumference, increased intracranial and growth retardation. Inclusion criteria for

SC

children with hydrocephalus were: 1) diagnosis of hydrocephalus based on Evans index > 0.30 [14]; or intracranial pressure >200mmH2O with clinical manifestations

MA NU

of high intracranial pressure - headache, vomiting, papilledema, unconsciousness; 2) patients underwent conventional MRI and 3D-T1WI FSPGR thin layer scanning; 3) patients had images from ventriculoscopy or CT ventriculography. The control group

Their inclusion criteria were: 1) no clinical manifestations

PT

based on age and gender.

ED

consisted of 15 normal children who were 1:1 matched with hydrocephalus children

of intracranial abnormalities, such as enlarged head circumference, intracranial

CE

hypertension, or growth retardation; 2) absence of clinical manifestation of

AC

neurological abnormalities; 3), no expansion of ventricle system based on conventional MR imaging and no intracranial lesions. This study was approved by the ethics committee of our institution. Patients or their guardians have signed a written informed consent prior to participation.

2.2 Instruments and reagents We used Singna EXCTTE 1.5T MR machines (GE, USA) and accompanying processing software (ADW 4.3) for MRI, OPTIMN CT660 (128T) machine (GE, USA) for CT, and PE-184A (Aesculap, Germany) for ventriculoscopy. CT contrast 6

ACCEPTED MANUSCRIPT agent was iopamidol, 10ml /vial, iodine concentration 200mg/ml, produced by Haibo

RI P

T

Lecco Xinyi Pharmaceutical (Shanghai, China)

SC

2.3 Methods

2.3.1 Scanning Sequence and Parameters

MA NU

Volume scanning was first performed with transverse three-dimensional (3D) fast spoiled gradient echo (3D-FSPGR) thin layer scan with the following parameters: section thickness 1mm, spacing 0mm, time of repetition (TR) 8.0 ms, time of echo

ED

(TE) 1.6 ms, excitation frequency 4. The parameters for conventional axial T1WI,

PT

T2WI, 3D-T1WI enhanced scan were as follows: T1WI, section thickness 5 mm, spacing 1 mm, TR2307 ms, TE10.6ms; T2WI: Fast Spin Echo (FSE) sequence,

CE

section thickness 5 mm, spacing 1 mm, TR4000ms, TE 103ms; 3D-T1WI enhanced

AC

scan: section thickness 1mm, spacing 0mm, TR8.2 ms, TE1.6 m. For sagittal 3D-CISS scan, the parameters were section thickness 2mm, spacing 0mm, TR47 ms, and TE1.5 ms.

2.3.2 Imaging Reconstruction and Virtual Endoscopic Imaging The imaging reconstruction was performed with the following procedures. (1) Transfer the original volume data into a workstation. (2) Use Smooth imaging Navigator software with black in white option; adjust threshold value (900~1200 for control group, 500 ~ 1100 for hydrocephalus group, the worse the CSF accumulation, 7

ACCEPTED MANUSCRIPT the lower the threshold) and visual fields to choose images for each subject; proceed from the third ventricle to the fourth ventricle and its outlets with flight through

RI P

T

function, to observe lumen dynamically; save important images. (3) Observe the cerebral ventricle inner wall from the distal end (bottom up) in the same way. Because

SC

the most common sites of obstruction in hydrocephalus are aqueduct, three outlets of the fourth ventricle, such as the foramen of magendie and foramina of Luschka. In

MA NU

this study we mainly observed these four locations.

2.3.3 CT ventriculography

ED

After general anesthesia, the needle was placed in the lateral ventricle through lateral

PT

horn of anterior fontanelle or skull puncture. Approximately 5ml of CSF were first released, followed by injection of 5ml contrast agent. The patient was kept in a

CE

Trendelenburg 30 degree position for 15 minutes, and then laid flat. Low dose (80 ~

AC

100kV/40mAs) CT scan was performed 6 hours after the contrast agent injection. If the ventricular system was not visible, another low-dose CT scan was performed at 12 hours of agent injection.

2.3.4 Ventriculoscopy The patient was placed in a supine position with head elevated, immobilized by skull clamp fixation. A hole was drilled at the location 1cm in front of the coronal suture and 3cm from midline. After penetration of lateral ventricles with brain puncture needle, the endoscope was inserted into the lateral ventricle through frontal cortex, 8

ACCEPTED MANUSCRIPT then to the third ventricle through Monro hole. Ventriculostomy was then performed

RI P

T

at the floor of the third ventricle to observe cerebral aqueduct.

All imaging analysis and comparison between MRVE and ventriculoscopy, CT

SC

ventriculography were performed by two experienced neuroradiologists.

MA NU

3. Results

3.1 Comparison of Results between MRVE and Ventriculoscopy

ED

Children in the control group (15cases): MRVE demonstrated lumina inner surface

PT

pattern in the aqueduct, the foramen of magendie and foramina of Luschka, characterized by smooth and uniform spiral lining along the long axis of the lumen

CE

and the evenly bright cavity wall. Ventriculoscopy of aqueduct revealed smooth

AC

surface and even color of inner wall (Fig. 1). The MRVE findings at three outlets of the fourth ventricle and aqueduct were comparable.

Typical signs of inner wall surfaces in obstructive hydrocephalus not induced by mass compression (11 cases): absence of the characteristics observed in MRVE of the control group, rough cavity wall，uneven brightness of ventricular inner surface that shows some super bright areas. Of them, 4 cases had a history of intracerebral hemorrhage and 5 cases had a history of encephalitis. For these 9 cases, ventriculoscopy revealed abnormal inner wall of aqueduct: multiple, brownish foreign 9

ACCEPTED MANUSCRIPT bodies or nodular lumps 3-5mm in diameter (Fig. 2). In 2 cases ventricle appeared slightly expanded, and patients’ age were less than 6 months. Medical history was

RI P

T

unclear. In one case ventriculoscopy was unremarkable, except the floor of the third ventricle was thick, intracranial pressure was 200mmH2O (Figure 3), and clinical

SC

symptoms improved significantly one month after ventriculostomy of the third

without other abnormal findings.

MA NU

ventricle. In another case ventriculoscopy showed only irregular edge of aqueduct

There were 4 cases with typical signs of mass compression-induced obstructive

ED

hydrocephalus: 1 with lateral ventricle posterior horn cysts, 1 with pineal cysts, 1 with

PT

cisterna magna cyst, and 1 case with the fourth ventricle tumor. The normal structure observed in MRVE of the control group were shown at aqueduct and outlets of the

CE

fourth ventricle, except in the areas suppressed by the mass . In ventroculoscopy the

AC

wall of cerebral ventricle remain smooth and lack of foreign body (Fig. 4);

3.2 Determine the Obstruction of Ventricular System and the Degree of Obstruction with MRVE The results of aqueduct MRVE were consistent with the findings of ventriculography or/and ventriculoscopy (15/15, 100%). There were 4 cases of complete occlusion of the aqueduct, where MRVE revealed disappearance of the normal structure observed in the control group (Figure 5a, b). There were 5 cases of incomplete occlusion of aqueduct (Figure 5c, d), and MRVE showed abrupt stenosis of aqueduct at concentric 10

ACCEPTED MANUSCRIPT distal end. In 2 cases there were no signs of aqueduct obstruction; and in 4 cases

RI P

T

aqueduct obstruction was due to tumor compression.

Another feature of of MRVE is to view the outlets of the fourth ventricle that were

SC

invisible to ventriculoscopy. In 5 patients who were suspected of the fourth ventricle outlet obstruction, MRVE showed complete occlusion of the fourth ventricle outlet

MA NU

which was later confirmed by CT ventriculography: expansion of proximal end and occlusion of distal end, no passage of contrast agent (Figure 5e, f, g).

ED

Among 15 hydrocephalus patients diagnosed with conventional MRI, enhanced MRI

PT

and 3D-CISS scan, 5 had incomplete obstruction of aqueduct, showing trumpet sign; 2 had obstructive membrane at the outlet of fourth ventricle; and 1 case had the fourth

AC

CE

ventricle tumor.

4. Discussion

Previous MRVE researches in ventricular system mainly focused on exploring the relationship of gross anatomy between the third ventricle and hypothalamus in order to prepare for the third ventriculostomy. There has been no report on the pathological changes of ventricular the control group inner surface [15-19], and study of its pathological changes during the development of hydrocephalus also relied on the microscopic or electron microscopic examination of animal autopsy [1, 20, 21]. Since 11

ACCEPTED MANUSCRIPT there are abundant gray matters around aqueduct where we are most concerned with for hydrocephalus, it is very difficult to obtain biopsy specimens from this area.

RI P

T

Moreover, observation of ventricle wall with ventriculoscopy is invasive. Thus it remains challenging to investigate the pathology of ventricular inner surface in

SC

hydracephalus patients with an non-invasive procedure. The primary objective of this study is to use MRVE to observe pathological changes of the ventricular inner surface,

MA NU

and compare the results with pathologic findings of ventriculoscopy, to evaluate the performance of this non-invasive procedure in identifying pathological changes of

ED

ventricular inner surface in hydrocephalus patients.

PT

4.1 The value of MRVE in the diagnosis of children with hydrocephalus – demonstrate the pathologic changes of ventricular inner surface

CE

Normal surface of ependyma has a monolayer of ciliated cuboidal epithelium, which

AC

under the electron microscope is smooth and absent of patchy tissue or cells. Consequently, normal ventricular inner surface has even surface and color [15]. In this study, we found that in the group of patients with obstructive hydrocephalus not induced by mass compression, 9 out of 11 cases demonstrated anomaly shadow on the inner wall during ventriculoscopy, indicating pathological changes of inner surface. This finding was consistent with the literature [4, 6-8, 12, 18, 22, 23] and the possible mechanisms may be as follows. Many literature [1, 20, 21, 23-26] reported the damage and repair of inner surface wall accompanied by microglia hyperplasia after encephalitis and intraventricular hemorrhage, which can in turn lead to CSF 12

ACCEPTED MANUSCRIPT circulation disorders and hydrocephalus. Multiple small brown spots are visible on the inner surface of ventricular system after hemorrhage, which may be the deposits of

RI P

T

hemoglobin [18]. Therefore, in our hydrocephalus patients who had encephalitis and cerebral hemorrhage, there had been a series of pathological changes to ventricular

SC

inner surface, resulted in the nodular shadow or hemoglobin deposit seen during ventriculoscopy. Furthermore, these shadows and deposits correspond with the

MA NU

superbright areas of MRVE. In the past, when MRVE was used for the diagnosis of colorectal lesions lumen bright areas were considered to be polyps [22]. Therefore, we

ED

believe MRVE can reflect pathologic changes of ventricular inner surface.

PT

In the group of patients with mass compression-induced obstructive hydrocephalus, all 4 patients were absent of anomaly shadow and their MRVE finds were similar to

CE

normal control. Results from MRVE are also consistent with the pathologic changes

AC

on the ventricular inner surface. This may be due to the fact that mild compression of cyst did not cause pathological changes of the ventricle inner surface under ventriculoscopy.

For 2 cases with slightly expanded ventricle, we performed ventriculoscopy because of their clinical symptoms and high intracranial pressure (200mmH2O). Anomaly shadow was not found on the ventricular wall, but MRVE indicated abnormal changes of ventricular inner surface, including rough surface, asymmetrical brightness of inner wall which were similar, but to a less degree, to the findings of MRVE in 13

ACCEPTED MANUSCRIPT hydrocephalus patients who had brown spots or nodular lumps in ventriculoscopy (compare Figure 3 and Figure 2). Since MRVE only observes the morphology, we

RI P

T

speculate that pathological changes may have occurred in ventricular inner surface. Literature also suggests that the pathologic change of ventricular inner surface in

SC

hydrocephalus is a continuous process, from visible destruction of ependymal structure as a result of mild ventricular dilation to severe ventricular dilation-induced

MA NU

gliosis under microscope or electron microscope. [18] For these 2 patients it was likely ventricular inner surface have changed [20], but not evident from ventriculoscopy. Therefore, MRVE may help detect early pathological changes of

PT

ED

ventricular inner surface.

In summary, different from previous MRVE researches that focused on the gross

CE

anatomy of ventricles [14, 16-19], we used MRVE to identify pathological changes of

AC

ventricular inner surface in children with hydrocephalus, and compared the results with findings from ventriculoscopy. This study helped us understand the ongoing changes of inner surface with the development of hydrocephalus. Although conventional MR and 3D-CISS are proven to be valuable in the diagnosis of hydrocephalus, they can only identify a case if obvious ventricular morphological changes have occurred or visible intraventricular obstructive membranes has appeared [4, 27], and they do not provide information on pathologic changes of inner surface as MRVE does. Therefore, MRVE technology can be a good complement to reveal pathological changes of ventricular inner surface in patients with hydrocephalus. 14

ACCEPTED MANUSCRIPT

4.2 The value of MRVE in the diagnosis of children with hydrocephalus –

RI P

T

determining whether there is obstruction and the degree of obstruction Using the same theory for the detection of stenosis or occlusion of the airway, MRVE

SC

can be used to detect the obstruction and determine the degree of obstruction in patients with hydrocephalus. The two applications have similar sensitivity and

MA NU

specificity [28-30]. In our study, the accuracy of detecting obstruction of the aqueduct with MRVE is 100%. The results for outlets of the fourth ventricle were also identical to CT ventriculography. Conventional MR examination and 3D-CISS are sensitive to

ED

morphological changes [4, 27]. In this study, only 8/15 cases had "trumpet" sign or

PT

obstructive membranes, indicating their low sensitivity in detecting obstruction. Also, these examinations cannot directly demonstrate the degree of obstruction. Therefore,

CE

MRVE has the advantages of high sensitivity, high accuracy, and ability to directly

AC

observe the obstruction and degree of obstruction over these modalities.

In addition, MRVE can cover a wide range, easily visualize the obstruction of ventricles below aqueduct, which is very meaningful for institutions lack of high quality ventriculoscopy.

This study has several limitations.

1), In 1 case (Figure 5a, b) multiple small brown

spots were not identified on MRVE. This is probably due to insignificant protrusion of these spots into the ventricular cavity, thus not sensitive to MRVE and cannot be 15

ACCEPTED MANUSCRIPT distinguished. 2) MRVE cannot be used for biopsy. Our study only had the results of gross pathology from ventriculoscopy, and did not have biopsy results from

RI P

T

microscopy or electron microscopy due to the difficulty of obtaining human biopsy

SC

specimen, to which future animal experiment may be beneficial.

In conclusion, MR virtual endoscopy is a non-invasive diagnostic approach for

MA NU

hydrocephalus. Supplementary to ventriculoscopy, it can provide intuitive visual imaging to clinicians, help observe pathological changes of ventricular inner surface, and detect obstruction and the degree of obstruction, thus should be considered in the

ED

diagnosis and treatment of hydrocephalus. However, the sample size of this study is

PT

limited, future study with larger patient population is necessary to confirm its

CE

findings.

AC

5. Ethical standards

This study was approved by the ethics committee of Shenzhen children’s Hospital. Patients or their guardians have signed a written informed consent prior to participation.

6. Conflict of interest The authors declare that they have no conflict of interest.

16

ACCEPTED MANUSCRIPT 7. References [1] Lattke M, Magnutzki A, Walther P, Wirth T, Baumann B. Nuclear factor kappaB

RI P

T

activation impairs ependymal ciliogenesis and links neuroinflammation to hydrocephalus formation. J Neurosci. 2012;32:11511-23.

SC

[2] Dincer A, Ozek MM. Radiologic evaluation of pediatric hydrocephalus. Childs Nerv Syst. 2011;27:1543-62.

MA NU

[3] Ben Achour N, Kraoua I, Rouissi A, Benrhouma H, Ben Youssef-Turki I, Jemel H, et al. [Congenital stenosis of interventricular foramina revealed by recurrent intracranial hypertension]. Neurochirurgie. 2013;59:93-6.

ED

[4] Shi J, Fu W, Wu Q, Zhang H, Zheng Z, Zhu J. Endoscopic third ventriculostomy

PT

associated 3D-construcive inference steady state MRI for obstructed hydrocephalus: a retrospective study. Clin Neurol Neurosurg. 2013;115:896-901.

CE

[5] Hilbert F, Wech T, Hahn D, Kostler H. Accelerated Radial Fourier-Velocity

AC

Encoding Using Compressed Sensing. Z Med Phys. 2013. [6] Bateman GA, Brown KM. The measurement of CSF flow through the aqueduct in normal and hydrocephalic children: from where does it come, to where does it go? Childs Nerv Syst. 2012;28:55-63. [7] Roth

J,

Ben-Sira

L,

Udayakumaran

S,

Constantini

S.

Contrast

ventriculo-cisternography: an auxiliary test for suspected fourth ventricular outlet obstruction. Childs Nerv Syst. 2012;28:453-9. [8] Summers RM, Shaw DJ, Shelhamer JH. CT virtual bronchoscopy of simulated endobronchial lesions: effect of scanning, reconstruction, and display settings and 17

ACCEPTED MANUSCRIPT potential pitfalls. AJR Am J Roentgenol. 1998;170:947-50.

RI P

gastrointestinal imaging. J Postgrad Med. 2010;56:109-16.

T

[9] McSweeney SE, O'Donoghue PM, Jhaveri K. Current and emerging techniques in

[10] Woo SR, Seo MW, Kim YH, Kwak HS, Han YM, Chung GH, et al. Extreme

SC

Duplication-type, Nonseparated Fenestration of the Basilar Artery in a Patient with Pontine Infarction: Confirmation with Virtual Arterial Endoscopy. J Clin Neurol.

MA NU

2006;2:74-7.

[11] van der Paardt MP, Stoker J. Magnetic resonance colonography for screening and diagnosis of colorectal cancer. Magn Reson Imaging Clin N Am. 2014;22:67-83.

ED

[12] Song HT, Huh YM, Kim S, Kim SJ, Suh JS. The usefulness of virtual MR

PT

arthroscopy as an adjunct to conventional MR arthrography in detecting anterior labral lesions of the shoulder. AJR Am J Roentgenol. 2009;192:W149-55.

CE

[13] Oishi M, Fukuda M, Takao T, Ishida G, Sato M, Fujii Y. [The utility of

AC

presurgical simulation of microvascular decompression by MR virtual endoscopy]. No Shinkei Geka. 2007;35:1087-95. [14] EVANS Jr WA. An encephalographic ratio for estimating ventricular enlargement and cerebral atrophy. Archives of Neurology and Psychiatry. 1942;47:931. [15] Krombach GA, Rohde V, Haage P, Struffert T, Kilbinger M, Thron A. Virtual endoscopy combined with intraoperative neuronavigation for planning of endoscopic surgery in

patients

with

occlusive

hydrocephalus

Neuroradiology. 2002;44:279-85. 18

and

intracranial

cysts.

ACCEPTED MANUSCRIPT [16] Burtscher J, Dessl A, Bale R, Eisner W, Auer A, Twerdy K, et al. Virtual endoscopy for planning endoscopic third ventriculostomy procedures. Pediatr

RI P

T

Neurosurg. 2000;32:77-82.

[17] Lemke AJ, Schurig-Urbaniak AM, Liebig T, Niehues SM, Haberl H, Lehmann

SC

TN, et al. [Virtual MR endoscopy of the ventricles prior to neurosurgical interventional endoscopy -- evaluation of different presentation techniques]. Rofo.

MA NU

2004;176:1106-13.

[18] Rohde V, Krombach GA, Struffert T, Gilsbach JM. Virtual MRI endoscopy: detection of anomalies of the ventricular anatomy and its possible role as a presurgical

ED

planning tool for endoscopic third ventriculostomy. Acta Neurochir (Wien).

PT

2001;143:1085-91.

[19] Burtscher J, Dessl A, Maurer H, Seiwald M, Felber S. Virtual neuroendoscopy, a

CE

comparative magnetic resonance and anatomical study. Minim Invasive Neurosurg.

AC

1999;42:113-7.

[20] Olopade FE, Shokunbi MT, Siren AL. The relationship between ventricular dilatation, neuropathological and neurobehavioural changes in hydrocephalic rats. Fluids Barriers CNS. 2012;9:19. [21] Okubo S, Strahle J, Keep RF, Hua Y, Xi G. Subarachnoid hemorrhage-induced hydrocephalus in rats. Stroke. 2013;44:547-50. [22] Schreyer, Feuerbach. Virtual endoscopy using MRI. Minim Invasive Ther Allied Technol. 2003;12:217-21. [23] Simard PF, Tosun C, Melnichenko L, Ivanova S, Gerzanich V, Simard JM. 19

ACCEPTED MANUSCRIPT Inflammation of the choroid plexus and ependymal layer of the ventricle following intraventricular hemorrhage. Transl Stroke Res. 2011;2:227-31.

RI P

T

[24] Roales-Bujan R, Paez P, Guerra M, Rodriguez S, Vio K, Ho-Plagaro A, et al. Astrocytes acquire morphological and functional characteristics of ependymal cells disruption

of

ependyma

in

hydrocephalus.

Acta

Neuropathol.

SC

following

2012;124:531-46.

MA NU

[25] Kleinman JT, Beslow LA, Engelmann K, Smith SE, Licht DJ, Ichord RN, et al. Evaluation of intraventricular hemorrhage in pediatric intracerebral hemorrhage. J Child Neurol. 2012;27:526-31.

ED

[26] Fadzli F, Ramli NM, Rahmat K, Ganesan D. Neonatal post-hemorrhagic

PT

hydrocephalus resulting in foraminal septae-radiological technique and surgical implications. Childs Nerv Syst. 2013;29:159-62.

CE

[27] Dincer A, Yildiz E, Kohan S, Memet Ozek M. Analysis of endoscopic third

AC

ventriculostomy patency by MRI: value of different pulse sequences, the sequence parameters, and the imaging planes for investigation of flow void. Childs Nerv Syst. 2011;27:127-35.

[28] Hara AK, Johnson CD, Reed JE, Ahlquist DA, Nelson H, MacCarty RL, et al. Detection of colorectal polyps with CT colography: initial assessment of sensitivity and specificity. Radiology. 1997;205:59-65. [29] Yunus M. Helical CT scan with 2D and 3D reconstructions and virtual endoscopy versus conventional endoscopy in the assessment of airway disease in neonates, infants and children. J Pak Med Assoc. 2012;62:1154-60. 20

ACCEPTED MANUSCRIPT [30] Davis CP, Ladd ME, Romanowski BJ, Wildermuth S, Knoplioch JF, Debatin JF. Human aorta: preliminary results with virtual endoscopy based on three-dimensional

AC

CE

PT

ED

MA NU

SC

RI P

T

MR imaging data sets. Radiology. 1996;199:37-40.

21

ACCEPTED MANUSCRIPT Figure Captions

RI P

T

Fig 1 MRVE, MRI and Ventriculoscopy of normal control, male, 3 years old. a), MRVE displayed smooth ependymal lines, even brightness of the ventricle inner wall

SC

surface, uniform change of the lumen diameter. b), Central sagittal image showed patency of aqueduct, and lack of signs of hydrocephalus. c),Ventriculoscopy of

MA NU

aqueduct demonstrated smooth surface and even color of inner wall.

Fig 2 Hydrocephalus patient with a history of encephalitis. Ventriculoscopy revealed

ED

abnormal aqueduct inner wall: multiple, brownish foreign bodies or nodular lumps

PT

a and b. male, 3 years and 5 months old, intracranial pressure 270 mmH2O. a), MRVE displayed patency and expansion of aqueduct, rough ventricular inner surface

CE

with uneven brightness, protruding-super-bright area (long arrows) on the inner wall,

AC

corresponding to brown spots and expansion of aqueduct (white arrow) on ventriculoscopy. b), Ventriculoscopy revealed aqueduct expansion (white arrow), a few brownish nodular lumps (black arrow) on the ventricular surface.

c and d, male, 1 year and 4 months old, growth retardation. c), MRVE indicated absence of the normal structure shown from the MRVE of the control group, rough cavity wall, uneven brightness, and several lumps on the surface shown as super bright spots (black arrows), corresponding to slightly elevated brown spots on ventriculoscopy; stenosis of lumen at distal end (white arrow).d), ventriculoscopy 22

ACCEPTED MANUSCRIPT revealed stenosis of aqueduct (white arrow), and shadows of brown spots (black

RI P

T

arrow) on the ventricular surface.

Fig 3 Patient with slightly expanded ventricle, growth retardation, and intracranial

SC

pressure above 200 mmH2O. male,1 year and 4 months old, a), MRVE displayed less smooth and uneven brightness of cavity wall in ventricular inner surface, and slightly

MA NU

narrow lumen at distal end. b), Sagittal image did not identify significant expansion at anterior crypt, neither did at other ventricles. c), Ventroculoscopy revealed smooth

ED

inner wall surface, no shadow of anomaly.

PT

Fig 4 Patient with typical signs of compression-induced obstructive hydrocephalus male, 1 year and 11 months old, ataxia. a), MRVE displayed the same structure as the

CE

the control group and visible shadow of large foreign body in the fourth ventricle

AC

(white arrow). b), Slight expansion at sagittal anterior crypt and large mass in the fourth ventricle (white arrow).

Fig 5 Patient with complete and incomplete occlusion of the aqueduct and/or the outlets of the fourth ventricle a and b, Male, 11 months old, growth retardation. a), MRVE showed occlusion at the opening of aqueduct (white arrow).b), Vetroculoscopy revealed complete occlusion of the aqueduct opening (white arrow), shadows of several brownish foreign bodies at the rear surface of the third ventricle (black arrow). 23

ACCEPTED MANUSCRIPT

c and d, male, 20 months old, increased head circumference, motor development

RI P

T

retardation, intracranial pressure 200mmH2O. c), MRVE displayed disappearance of the normal structure of aqueduct observed in the control group ，rough cavity wall，

SC

and complete occlusion of the aqueduct. (white arrow). d), Ventriculoscopy revealed stenosis of aqueduct (white arrow), no shadow of foreign body on the ventricular

MA NU

surface.

e, f and g, female, 3 years old, growth retardation. e), MRVE showed dilated outlets

ED

of the fourth ventricle, disappearance of the normal structure,rough surface,

PT

asymmetrical brightness, and several lumps;

visible dilation of lumen at proximal

end and occlusion at distal end (white arrows); f and g), CT ventriculography revealed

CE

no passage of contrast agent through the foramina of Luschka into prepontine cistern

AC

six hours after the injection (white arrow).

24

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

Fig. 1

25

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

Fig. 2

26

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

SC

Fig. 3

27

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA NU

Fig. 4

28

AC

CE

PT

ED

MA NU

SC

RI P

T

ACCEPTED MANUSCRIPT

Fig. 5

29