[1] 14w Evaluation of CSF flow metrics in patients with communicating hydrocephalus and idiopathic intracranial hypertension

[1] 267w PurposeTo search for CSF dynamics of idiopathic intracranial hypertension (IIH) and communicating hydrocephalus and any correlation between MRI findings, CSF metrics and CSF opening pressure in IIH. Materials and methods Healthy subjects (30) and subjects with IIH (29) and high-pressure communicating hydrocephalus (43) were enrolled. Nonparametric Kruskal-Wallis test (p = 0.05) was used to compare three groups, Mann-Whitney U test with Bonferroni correction to compare two groups (p = 0.016). Correlation of MRI findings of IIH with CSF metrics and CSF opening pressure was analyzed by Spearman's Rank correlation coefficient (p = 0.05). ResultsIn IIH, no correlation between MRI findings and aqueductal stroke volume (ASV) but statistically significantly CSF opening pressure in the presence of transverse sinus compression was noted. Comparing with healthy subjects, ASV was nonsignificantly lower and standardized diastolic and sum and difference of systolic and diastolic flow durations were statistically significantly lower. Comparing with hydrocephalus, the width of prepontine cistern (PPC)/the width of aqueductus sylvii (AS) was significantly higher and other CSF metrics with standardized systolic and sum of systolic and diastolic flow durations were significantly lower. In hydrocephalus, ASV and peak velocities were significantly higher. Compared with normal group, PPC/AS and reverse/forward flow duration were significantly lower and other CSF metrics were significantly higher. Conclusion In hydrocephalus, significant increase in ASV and peak velocities were noted. In IIH, CSF opening pressure was statistically significantly high in the presence of transverse sinus compression and standardized diastolic flow durations were statistically significantly short that are probably effects of increased impedance of CSF flow against increased intracranial pressure and unchanged or even decreased intraventricular CSF volume.

[1] 164w Hydrocephalus is a complex disorder system caused by various etiologies and defined as dilatation of the ventricular and an increase in volume of CSF. It could be communicating or noncommunicating depending on CSF can exit the ventricular system or not, respectively. Communicating hydrocephalus could be with obstruction to CSF absorption secondary to subarachnoid bleeding, leptomeningeal carcinomatosis or infective meningitis or could be without obstruction to CSF absorption such as in normal pressure hydrocephalus, hydrocephalus ex-vacuo or colpocephaly secondary to parenchymal volume loss or in choroid plexus papillomas due to overproduction of CSF. Most common imaging findings of hydrocephalus are: ventriculomegaly (Evans' index > 0.3), enlargement of the third ventricular recesses and lateral ventricular horns, decreased mamillopontine distance and frontal horn angle, thinning and elevation of the corpus callosum, normal or narrowed cortical sulci, periventricular white matter hyperintensities due to interstitial edema and aqueductal flow void phenomenon in T2 W images. In CSF flow studies, dilated aqueductus sylvii and increased ASV may be seen [1,2].

[2] 158w Pseudotumor cerebri or IIH is a disorder of unknown etiology that is characterized by elevated intracranial pressure, usually occurs in obese women in the childbearing years [3][4][5]. Patient with IIH maintains an alert and oriented mental state with no localizing neurological findings but headache and visual problems. There is no evidence of deformity or obstruction of the ventricular system, and neurodiagnostic studies are otherwise normal except for increased cerebrospinal fluid pressure (greater than 200 mm of water in the non-obese and probably greater than 250 mm of water in the obese patient). MRI findings of IIH are widening of perioptic subarachnoid space, optic nerve tortuosity, posterior scleral flattening, preliminary enhancement of optic nerves, empty sella, tonsillar herniation, perimesencephalic cistern obliteration, shrinkage of ventricles, enlargement of Meckel's cave and bilateral transverse sinus compression [6]. In the literature, only one study was present detailing CSF flow metric in IIH suggesting statistically decreased peak flow velocity in untreated IIH patients [4].

[3] 135w With recent advances on MRI, MRI is not only beneficial in the diagnosis of CSF-related diseases, but also helps clinicians in planning the treatment and follow-up of patients. New MRI sequences including phase-contrast magnetic resonance imaging (PC-MRI), 3D-heavily T2W (3D constructive interference in steady state) (3D-CISS) and 3D sampling perfection with application-optimized contrasts using different flip angle evolutions (3D-SPACE) enable evaluation of CSF-related pathologies with higher sensitivity and specificity. Many parameters in CSF-related disorders have been studied [7]. There are various factors that influence CSF dynamic such as ventricular size, aqueductal diameter, brain compliance and cerebral blood flow [8]. This study is retrospective case-control observational study, and our purpose is to evaluate CSF dynamics of IIH and communicating hydrocephalus and search for any correlation between MRI findings, CSF metrics and CSF opening pressure in IIH.

[1] 42w In subjects with communicating hydrocephalus, mean age was 29.12 years (age range 11-78 years), in subjects with IIH, mean age was 37.28 years (age range 18-57 years), and in healthy volunteers, mean age was 36.57 years (age range 20-65 years) (Table 1).

[2] 255w Average (mean) ASV values were measured as 23.61 (SD ± 12) µl in healthy subjects, 21.1 (SD ± 14) µl in IIH and 194.8 (SD ± 97, 6) µl in communicating hydrocephalus (Fig. 4). Average volume flow in one cardiac cycle was measured as 4.19 (SD ± 2.1) ml/min in healthy subjects, 3.7 (SD ± 2.4) ml/min in IIH and 34.1 (SD ± 17.4) ml/min in communicating hydrocephalus. Average peak flow velocities were measured as 4.97 (SD ± 1.61) cm/s in healthy subjects, 4.8 (SD ± 3.76) cm/s in IIH and 12.50 (SD ± 4.37) cm/s in communicating hydrocephalus (Fig. 5). When comparing IIH with healthy subjects, standardized diastolic flow time [median values in IIH, 395 ms/ healthy group, 484 ms; (p = 0.004)] and difference between standardized diastolic and systolic flow time [median values in IIH, 206 ms/healthy group, 289 ms) (p = 0.010)] were significantly short in IIH. Nonstandardized diastolic flow time showed statistically borderline difference [median values in IIH, 363 ms/healthy group, 427 ms; (p = 0.020)]. When the sum of standardized systolic and diastolic flow times was taken into consideration, total flow time was also significantly short in IIH compared to healthy subjects (median values in IIH, 542 ms (minimum 360 ms, maximum 755 ms)/healthy group, 636 ms (minimum 374 ms, maximum 821 ms) (p = 0.011) ( 0.00 5.00 10.00 15.00 20.00 25.00 0 1 0 2 0 3 0 4 0 5 0 cm/sn n PSV IIH HEALTHY SUBJECTS HYDROCEPHALUS Fig. 5 Peak flow velocity values of three groups

[3] 498w Table 2 CSF metrics of three groups Group PPS/AS ASV Peak flow velocity (cm/sn) Forward volume 10 -3 ml Reverse volume 10 -3 ml Net forward volume 10 -3 ml Average flow over range 10 -3 ml/sn Average velocity 10 -3 cm/sn Average area 10 -2 cm 2 Average reverse velocity duration (ms) Average reverse velocity 10 -3 cm/sn Average forward velocity duration (ms) IIH N 29 29 29 29 29 29 29 29 29 28 28 28 Mean 3.7552 21.12 -.38 18.10 24.06 -6.00 -11.89 -282.20 5.38 378.03 .06 169.37 SD .88917 14.02 6.16 13.59 17.48 13.71 18.59 365.68 3.99 69.45 .04 30.38 Median 3.7500 15.00 -2.11 12.00 21.00 -8.00 -13.00 -313.00 4.50 363.00 .05 172.86 Minimum 1.95 6.00 -7.73 4.00 0.00 -32.00 -47.00 -931.00 .42 239.00 0.00 110.00 Maximum 5.00 69.50 21.00 59.00 80.00 30.00 30.00 623.00 17.80 510.00 .19 235.00 Normal N 30 30 30 30 30 30 30 30 30 30 30 30 Mean 4.0667 23.61 -.63 17.56 29.93 -12.26 -18.06 -464.86 4.08 421.50 .07 171.67 SD .89763 12.01 5.27 10.46 14.49 8.79 13.08 342.63 1.69 69.10 .03 29.41 Median 4.0000 21.25 -2.58 16.50 26.00 -9.50 -14.50 -461.50 3.60 427.00 .06 170.00 Minimum 2.50 4.00 -7.14 1.00 7.00 -31.00 -46.00 -1300.00 1.70 267.00 .01 126.97 Maximum 6.00 54.50 9.41 44.00 65.00 6.00 9.00 369.00 9.80 540.00 .17 242.00 Hydrocephalus N 43 43 43 43 43 43 43 43 43 43 43 43 Mean 1.9542 194.88 -5.35 153.16 233.41 -80.25 -97.86 -287.11 2.16 385.02 .60 175.13 SD .74594 97.67 12.23 94.37 135.42 124.37 197.85 643.35 1.20 98.60 .33 33.83 Median 2.0000 180.00 -11.26 141.00 205.00 -104.00 -144.00 -187.00 2.07 413.00 .56 174.00 Minimum .71 21.00 -18.73 14.00 16.00 -309.00 -454.00 -2550.00 .50 152.00 .07 92.00 Maximum 4.25 404.00 18.14 405.00 540.00 285.00 418.00 784.00 5.55 529.00 1.38 240.00 Group Average forward velocity 10 -3 cm/sn Average forward velocity/average reverse velocity Heart rate (FV + RV)/680*60 (ml/min) Reverse velocity duration × heart rate/60 Forward velocity duration × heart rate/60 Average reverse-forward duration (ms) (Reverseforward duration) × heart rate/60 Average reverse duration/ forward velocity duration Average reverse + forward duration (ms) Reverse duration × heart rate/60 × reverse volume 10 -3 ml IIH N 28 27 26 29 26 26 28 26 28 28 25 Mean .10 1.79 59.73 3.72 374.99 167.83 208.66 207.16 2.29 547.41 9943.66 SD .07 1.59 9.69 2.47 100.74 42.26 72.38 85.81 .57 79.09 8360.74 Median .07 1.32 59.50 2.64 395.40 155.40 200.11 206.79 2.25 538.26 7617.60 Minimum .02 .53 43.00 1.06 211.60 104.50 79.00 62.59 1.42 399.00 1666.00 Maximum .30 7.98 73.00 12.26 510.00 274.17 322.48 335.59 3.58 697.52 standardized systolic and diastolic flow time was evaluated, total flow time was statistically significantly short in IIH compared to subjects with hydrocephalus (median values in IIH, 542 ms (minimum 360 ms, maximum 755 ms)/hydrocephalus, 618 ms (minimum 324 ms, maximum 807 ms) (Figs. 4, 5, 6). A statistically significant difference was not determined in other CSF flow metrics between two groups (Table 2).

[4] 147w In patients with IIH, there was no correlation of empty sella, optic nerve sheath edema, posterior scleral flattening, perimesencephalic cistern obliteration, shrinkage of ventricles, enlargement of Meckel's cave, tonsillar herniation, transverse sinus compression and PPC/AS ratio with ASV values. Among conventional MR and MR venography findings including PPC/AS ratio and ASV, there was only statistically significant difference in CSF opening pressure between IIH patients with transverse sinus compression and IIH patients without transverse sinus compression. In all subjects with IIH (except one subject in whom MR venography was not diagnostic due to technical failure), in subjects with transverse sinus compression (17/28) median CSF opening pressure was measured as 34.5 cm H 2 O (minimum 22, maximum 60 cm H 2 O); in subjects without transverse sinus compression (11/28) this value was 24 cm H 2 O (minimum 21, maximum 40 cm H 2 O) (p = 0.05).

[1] 87w CSF flow metrics by means of flow-sensitive MRI sequences have been investigated in normal subjects and in different pathologies such as high-pressure communicating hydrocephalus, normal pressure hydrocephalus (NPH) and IIH [9]. Aqueductal stroke volume (ASV), a flow metric, is the average of forward flow volume during systole and reverse flow volume during diastole in each cardiac cycle. Baledent et al. [7] reported higher ASV values in communicating hydrocephalus compared to normal subjects (200 to 50 µl). In the literature, ASV was reported as 30-50 µl in normal

[2] 244w Table 2 (continued) Group Average forward velocity 10 -3 cm/sn Average forward velocity/average reverse velocity Heart rate (FV + RV)/680*60 (ml/min) Reverse velocity duration × heart rate/60 Forward velocity duration × heart rate/60 Average reverse-forward duration (ms) (Reverseforward duration) × heart rate/60 Average reverse duration/ forward velocity duration Average reverse + forward duration (ms) Reverse duration × heart rate/60 × reverse volume 10 -3 ml Normal N 30 30 30 30 30 30 30 30 30 30 30 Mean .10 1.41 64.00 4.19 452.84 183.62 249.82 269.21 2.48 593.17 13506.19 SD .06 .60 9.321 2.09 107.77 42.98 60.77 81.33 .41 87.11 7234.713 Median .09 1.39 67.50 3.79 484.39 190.97 254.00 289.01 2.52 612.50 12309.46 Minimum .01 .47 44.00 .71 253.65 105.60 140.03 133.03 1.85 393.97 1775.55 Maximum .30 3.91 77.00 9.62 612.00 274.27 354.00 424.80 3.38 743.15 30288.92 Hydrocephalus N 43 43 43 43 43 43 43 43 43 43 43 Mean .89 1.76 68.37 34.11 422.48 196.43 209.88 226.05 2.22 560.16 100016.21 SD .57 1.54 17.46 17.42 96.66 49.40 89.13 92.01 .56 117.43 63274.18 Median .75 1.25 66.00 31.76 432.50 203.55 219.00 230.78 2.15 583.00 87578.83 Minimum .08 .46 35.00 3.62 207.77 98.58 -36.00 -63.60 .83 244.00 6650.67 Maximum 2.38 9.55 136.00 71.12 593.23 365.70 400.00 413.33 4.15 734.00 238828.20 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 0 1 0 2 0 3 0 4 0 5 0 n PPC/AS IIH HEALTHY SUBJECTS HYDROCEPHALUS Fig. 6 PPC/AS ratio values of three groups

[3] 126w healthy subjects [10][11][12][13]. In our study, normal ASV value was found as 23.61 µl (SD ± 12). In a study with normal volunteers, Lee et al. reported that the mean peak systolic velocities showed a tendency to increase from the superior (pars anterior) to the inferior (pars posterior) aqueduct, ranging from 3.30 to 4.08 cm/s without statistically significant differences. They found highest peak velocities in the caudal third of the AS (pars posterior), which is narrowest part of the AS [14]. We observed mean peak velocities 4.97 (SD ± 1.61) cm/sn in healthy subjects, 4.80 cm/sn in IIH and 12.50 cm/sn in hydrocephalus. However, location of axial views of PC-MRI in our study was through ampulla of AS, which is widest and middle-third area of AS.

[4] 80w Normal CSF flow at the level of AS and foramen magnum is biphasic due to pump mechanism. Systolic expansion of intracranial vessels and brain parenchyma results in CSF movement from ventricular system toward the spinal canal (craniocaudal flow, systolic CSF flow, forward flow), while diastolic contraction of intracranial vessels and brain parenchyma results in reduction in intracranial pressure and secondary CSF movement from spinal region to subarachnoid space and ventricular system within the brain (caudocranial flow, diastolic flow, reverse flow).

[5] 166w In this study, we measured diastolic flow time in IIH and healthy subjects. In IIH comparing with healthy subjects, standardized diastolic flow time (median value in IIH, 395 ms/healthy subjects, 484 ms) and difference between standardized diastolic and systolic flow time (median value in IIH, 206 ms/healthy subjects, 289 ms) and sum of standardized systolic and diastolic flow times (median value in IIH 542 ms/healthy subjects 636 ms) were found statistically significantly short. Comparing the sum of standardized systolic and diastolic flow times among three groups, it was statistically shortest in IIH (median value in healthy subjects 636 ms, hydrocephalus 618 ms, IIH 542 ms), but not statistically significant difference noted between hydrocephalus and healthy subjects. This shortening in IIH is caused by shortening of diastolic flow duration. We thought that impedance of caudocranial flow against increased intracranial pressure in IIH results in shortening in diastolic flow time. In IIH, we also observed abnormal CSF flow pulsatile resulting from attempt to flow against the increased pressure.

[6] 205w CT and MRI findings in patients with IIH are dilated optic nerve sheath, empty sella, posterior scleral flattening, prelaminar optic nerve enhancement, vertical tortuous appearance of orbital optic nerves, enlargement of Meckel cave and rarely tonsillar herniation. The presence of slit-like ventricles is controversial; ventricles could be small or normal [4]. In a study of Diyvata et al., MRI findings were found similar in both IIH and secondary intracranial hypertension [15]. They reported that optic nerve head protrusion and posterior scleral flattening were significantly associated with IIH with no statistically significant difference found in the occurrence of rest of the findings. In another study of IIH, Riggeal et al. [16] assessed presence and percentage of transverse sinus stenosis and correlation between the degree of stenosis and the clinical course and CSF opening pressure. They reported bilateral transverse sinus stenosis in 46 patients among 51 patients. In 71% of patients with transverse sinus stenosis, stenosis ratio was greater than 50%. They found no correlation between the degree of stenosis and the clinical course. CSF opening pressure was not found associated with location or degree of transverse sinus stenosis. They suggested that clinical features, not the degree of stenosis, should be used to determine management in IIH.

[7] 254w In our study, we compared CSF opening pressure and CSF flow metrics with cranial MRI and MR venography findings. There was no significant correlation between CSF opening pressure and CSF flow metrics. However, there was a statistically significant high CSF opening pressure in IIH patients with transverse sinus compression, compared to the ones without transverse sinus compression. Rohr et al. [3] reported that degree of transverse sinus compression was increased with higher CSF opening pressures. This finding is compatible with our study, but the study of Rohr et al. consisted of subjects with secondary intracranial hypertension. Bono et al. [17] reported that in 64% of patients with IIH, CSF pressure normalized during medical treatment despite transverse sinus stenosis persisted in all. Contrary to our study and study of Rohr et al., they suggested that there was no direct relationship between the caliber of transverse sinus and CSF opening pressure. Metellus et al. [18] reported a case of intracranial hypertension with 40 mm Hg CSF opening pressure treated by endovascular stent placement in the right transverse sinus. After stenting, the patient's symptoms improved, papilledema disappeared, MRI findings returned to normal, and CSF pressure on lumber puncture was lowered to 11 mmHg. They hypothesized that transverse sinus compression was the cause rather than result. In contrast to Metellus et al., we suggest that transverse sinus compression might be result rather than cause since we encountered higher CSF opening pressure in IIH patients with transverse sinus compression in comparison with IIH patients with no transverse sinus compression.

[8] 212w Akay et al. [4] found no statistically significant difference in width of AS between IIH patients and control group. In our study, in comparison with AS area and PPC/ AS ratio, there was no significant difference between patients with IIH and healthy subjects but a statistically significant difference was found between hydrocephalus and other two groups. In hydrocephalus, there was a significant increase in AS area. Median AS areas were measured as 3.60 mm 2 in healthy subjects, 4.5 mm 2 in IIH and 20.7 mm 2 in hydrocephalus. This led to a statistically significant decrease in PPC/AS ratio in hydrocephalus compared to two other groups. Median PPC/AS ratio was measured as 4 in healthy subjects, 3.75 in IIH and 2 in hydrocephalus. Abbey et al. [19] reported wider AS areas in patients with communicating hydrocephalus before ventriculoperitoneal shunt surgery compared to normal group. In addition, Abbey et al. found significantly higher ASV, peak systolic and diastolic flow velocities in patients with communicating hydrocephalus compared to normal group. In the present study, we found statistically significant increase in ASV, peak flow velocity, forward flow volume, reverse flow volume, net forward flow volume, forward flow velocity, reverse flow velocity and average flow volume in one cardiac cycle in hydrocephalus, compared to healthy subjects.

[9] 98w High ASV values in hydrocephalus are due to increased CSF volume and increased AS area. Chiang et al. reported a strong correlation between ASV and total ventricular volume and width of third ventricle. They found a linear correlation between ASV and AS area [20]. In our study, standardized forward flow time (craniocaudal flow) was increased due to the increased CSF volume in hydrocephalus. Median standardized forward flow time was measured as 203 ms in hydrocephalus, 155 ms in IIH and 190 ms in healthy subjects. However, only statistically significant difference was found between patients with hydrocephalus and IIH.

[10] 248w In comparison with ASV values of all three groups, a significant increase in ASV was detected in hydrocephalus, whereas no significant difference between IIH and healthy subjects (median ASV 180 µl in hydrocephalus, 15 µl in IIH and 21.2 µl in healthy subjects). The average flow volume in one cardiac cycle showed similar ranking (median value 31.76 ml/min in hydrocephalus, 2.6 ml/min in IIH, and 3.7 ml/min in healthy subjects). We observed lowest ASV and average flow volume in one cardiac cycle in IIH without statistically significant difference in comparison with healthy subjects. ASV shows the volume of CSF passing through AS, and it has a positive correlation with area of AS and volume of CSF within the lateral and third ventricles. In spite of the fact that AS lumen area was larger in IIH compared to healthy subjects (4.5 mm 2 vs 3.6 mm 2 respectively), the absence of expected change in ASV may be explained by unchanged or even decreased intraventricular CSF volume in IIH. Alperin et al. reported normal intraventricular volume, but significantly increased extraventricular CSF volume in a study performed in obesity-associated IIH [5]. This study partly supports our hypothesis that unchanged or even decreased intraventricular CSF volume in IIH. In our study, standardized diastolic flow time and sum of standardized systolic and diastolic flow times were significantly short in IIH compared to other groups. We thought that these findings were expected effects of decreased intraventricular CSF volume as well as increased intracranial pressure.

[1] 225w CSF flow dynamic is so complex, and pathways of CSF production and absorption are still speculative. In this study, we evaluated CSF flow dynamics in normal subjects, IIH and communicating hydrocephalus. We tried to explain the pathophysiology of diseases by using CSF flow metrics. In communicating hydrocephalus, increase in ASV and peak flow velocity are striking findings. We suggested that increase in ASV in communicating hydrocephalus is due to increase in intraventricular CSF volume and wider AS lumen area. In IIH, we found statistical significant increased CSF opening pressure in subjects with transverse sinus compression compared to ones without transverse sinus compression. We suggested that transverse sinus compression is the result rather than the cause of IIH. Although there was no statistically significant difference in ASV between normal and IIH groups, IIH group has lowest ASV values, comparing with other groups. In IIH, among CSF flow metrics, standardized diastolic flow time and sum and difference of standardized systolic and diastolic flow times were all significantly short. This mainly comes from shortening in diastolic flow duration that is probably effects of increased impedance of CSF flow against increased intracranial pressure and unchanged or even decreased intraventricular CSF volume. In conclusion, CSF flow dynamic is not fully understood; further, more detailed, larger-scale studies should be performed to reveal the role and variability of CSF flow in diseases.

[1] 221w Twenty-nine patients (average age 37.28, M/F 2/27) (age range 18-57 years) with a diagnosis of IIH and 43 patients (average age 29.12, M/F 23/20) (age range 11-78 years) with a diagnosis of high-pressure communicating hydrocephalus based on standard clinical and laboratory criteria, referred from our neurology department, were included. Cases with normal pressure hydrocephalus, hydrocephalus ex-vacuo or colpocephaly secondary to parenchymal volume loss or hydrocephalus due to overproduction such as in choroid plexus papilloma were all excluded. Control group included 30 healthy volunteers (average age 36.57, M/F 25/5) (age range 20-65 years), with normal neurological examination, clinical and laboratory findings. All CSF flow studies of the subjects with IIH and communicating hydrocephalus were performed before their treatment. In subjects with IIH, lumbar puncture for CSF opening pressure was performed after CSF flow and MRI examination in 12 h. Subjects with IIH underwent additionally contrast-enhanced MR venography examination to exclude venous sinus thrombosis. For all patients in this study, a detailed history, neurological examination and laboratory findings were retrospectively evaluated in our hospital via computer-based archive system. Healthy volunteers also underwent CSF flow study. The procedures used were in accordance with the guidelines of the Declaration of Helsinki on human experimentation. The study protocol was approved by our institutional ethical committee. All subjects were fully informed and gave their written informed consent.

[2] 269w MRI was performed on a 1.5 Tesla system (Avanto; Siemens Medical Solution, Erlangen, Germany) using head coil. The quantitative evaluation of CSF flow was performed with images obtained in the axial plane with the two-dimensional (2D) Q FLOW phase-contrast MR angiography technique. Axial views (through plane views) of PC-MRI passed through ampulla of AS, which is widest and middle-third area of AS. The PC-MRI duration for each patient was approximately 5 min (Fig. 1). Midline sagittal, coronal and axial T1 W preliminary images were first obtained. "Mean modulus," "magnitude of complex difference" and "directional phase difference" images perpendicular to the cerebral aqueduct in the sagittal plane were then obtained at the semi-axial plane. For images in the axial plane, the parameters used for the 14-30 cardiac phase sections according to heart rate were: TR: 31.25 ms, TE: 8.06 ms, slice thickness: 3 mm, NSA: 1, FOV: 16 × 10 cm, matrix 128 × 256 and flip angle, 10°. Cardiac triggering was performed retrospectively with MR compatible electrodes (Kendall, Arbo, Tyco International, Neustadt, Germany). Flow sensitivity (venc) was determined as 20 cm/s. Flow in the caudocranial direction (diastolic or reverse flow) was identified as negative, and flow in the craniocaudal direction (systolic or forward flow) was identified as positive. After PC-MRI, sagittal T2 CISS and T2 SPACE sequences were also obtained. Parameters used for 3D-T2 CISS were: slice thickness: 1 mm, FOV: 200 mm, matrix 290 × 320, TR: 6.06 ms, TR: 2.61 ms and flip angle 70°. Parameters used for 3D-T2 SPACE were: slice thickness: 1 mm, FOV: 240 mm, matrix 231 × 256, TR: 2500 and TR: 501.

[3] 202w The images obtained from all subjects (subjects with IIH, subjects with hydrocephalus and healthy subjects) were evaluated by two separate radiologists (AA and TFY) using the Siemens user console (Argus software, Siemens, Erlangen, Germany). Any type of disputes was resolved through consensus. On midline sagittal T2 CISS images, at the midpontine level, width of prepontine cistern (PPC) and width of cerebral aqueductus sylvii (AS) were measured and PPC/ AS ratio was calculated. PC-MRI through plane (axial) views were used for CSF flow quantification. Axial image with a widest flow-related signal through aqueduct sylvii was chosen. Flow contours were drawn by a region of interest (ROI), and ROI was copied over entirely in axial phase images obtained during one cardiac cycle (Fig. 2). Velocity versus time, peak flow velocity versus time (Fig. 3), flow versus time and net flow versus time curves of the flow passing through the AS during one cardiac cycle were obtained. Peak and average velocity (cm/s), forward and reverse flow In IIH patients, accompanying conventional MRI and MR venography findings including empty sella, optic nerve sheath edema, posterior scleral flattening, perimesencephalic cistern obliteration, shrinkage of ventricles, enlargement of Meckel's cave, tonsillar herniation and transverse sinus compression were also noted.

[4] 79w All statistical analyses were performed using SPSS for Windows version 23.0 software package (IBM Corp., New York, NY; formerly SPSS Inc., Chicago, IL). Subjects were evaluated in three separate groups: Group with communicating hydrocephalus (n = 43), group with IIH (n = 29) and group with healthy volunteers (n = 30). For each three groups, mean ± standard deviation of frequently used CSF flow metrics (ASV, average volume flow in one cardiac cycle, and peak flow velocities) was calculated.

[5] 24w First, distribution of variables from three groups was evaluated by using Shapiro-Wilk normality test. For each group, median, maximum and minimum values were determined.

[6] 71w Since distribution of variables among three different groups was so wide, nonparametric test (Kruskal-Wallis variance analysis test) was used for three-group comparison. A p value < 0.05 was accepted as statistically significant. Mann-Whitney U test was used for comparison of variables between two groups. In this evaluation, for all affects, by performing Bonferroni correction the level of significance was determined as 0.016. (Normal p value was divided by group number 0.05/3.)

[7] 47w In subjects with IIH, correlations of conventional MRI and MR venography findings with ASV values and CSF opening pressures were evaluated. Spearman's Rank correlation coefficient, which is a nonparametric test, was used for correlations of continuous variables. A p value < 0.05 was accepted as statistically significant.