Cerebrospinal Fluid Pressure Reduction Results in Dynamic Changes in Optic Nerve Angle on Magnetic Resonance Imaging

Background: Optic nerve sheath tortuosity is a previously reported, but incompletely characterized, finding in idiopathic intracranial hypertension (IIH). We hypothesized that optic nerve angle (ONA), as a quantitative measure of tortuosity, would change dynamically with cerebrospinal fluid (CSF) pressure status of patients with IIH immediately before and after lumbar puncture (LP). Methods: Consecutive patients with suspected IIH referred for MRI and diagnostic LP were prospectively enrolled in this single institution, institutional review board-approved study. Each patient underwent a pre-LP MRI, diagnostic LP with opening pressure (OP) and closing pressure (CP), and then post-LP MRI all within 1 session. Sagittal and axial ONAs were measured on multiplanar T2 SPACE images by 2 neuroradiologists on pre- and post-LP MRI. Effects of measured pressure and CSF volume removal on changes in ONA were analyzed as was interrater reliability for ONA measurement. Results: Ten patients with IIH were included {all female, median age 29 (interquartile range [IQR] 25-32)}. All patients had elevated OP (median 37, IQR 34-41 cm H2O), and significantly reduced CP (median 18, IQR 16-19 cm H2O, P < 0.001) after CSF removal (IQR 13-16 mL). Within patients, mean ONAs (sagittal and axial) were significantly lower before (162 ± 9°, 163 ± 10°) than after (168 ± 7°, 169 ± 5°) LP (P = 0.001, 0.008, respectively). Interrater reliability was higher with sagittal ONA measurements (0.89) than axial (0.72). Conclusions: ONA changes with short-term CSF pressure reduction in patients with IIH, establishing optic nerve tortuosity as a dynamic process related to CSF status.

Original Contribution Cerebrospinal Fluid Pressure Reduction Results in Dynamic Changes in Optic Nerve Angle on Magnetic Resonance Imaging Ranliang Hu, MD, John Holbrook, MD, Nancy J. Newman, MD, Valerie Biousse, MD, Beau B. Bruce, MD, PhD, Deqiang Qiu, PhD, John Oshinski, PhD, Amit M. Saindane, MD Background: Optic nerve sheath tortuosity is a previously reported, but incompletely characterized, finding in idiopathic intracranial hypertension (IIH). We hypothesized that optic nerve angle (ONA), as a quantitative measure of tortuosity, would change dynamically with cerebrospinal fluid (CSF) pressure status of patients with IIH immediately before and after lumbar puncture (LP). Methods: Consecutive patients with suspected IIH referred for MRI and diagnostic LP were prospectively enrolled in this single institution, institutional review board-approved study. Each patient underwent a pre-LP MRI, diagnostic LP with opening pressure (OP) and closing pressure (CP), and then post-LP MRI all within 1 session. Sagittal and axial ONAs were measured on multiplanar T2 SPACE images by 2 neuroradiologists on pre- and post-LP MRI. Effects of measured pressure and CSF volume removal on changes in ONA were analyzed as was interrater reliability for ONA measurement. Results: Ten patients with IIH were included {all female, median age 29 (interquartile range [IQR] 25-32)}. All patients had elevated OP (median 37, IQR 34-41 cm H2O), Department of Radiology and Imaging Sciences (RH, DQ, JO, and AMS), Emory University School of Medicine, Atlanta, Georgia; Northside Radiology Associates (JH), Atlanta, Georgia; and Department of Ophthalmology (NJN, VB, BBB), Emory University School of Medicine, Atlanta, Georgia. B. B. Bruce is a medicolegal consultant for Bayer and individual litigants on the topic of idiopathic intracranial hypertension. V. Biousse and N. J. Newman are consultants for GenSight Biologics. They are supported in part by an unrestricted departmental grant (Department of Ophthalmology) from Research to Prevent Blindness, Inc, New York, by NIH/NEI core grant P30-EY06360 (Department of Ophthalmology, Emory University School of Medicine), and by NIH/NINDS (RO1NSO89694). N. J. Newman is a consultant for Santhera Pharmaceuticals. J. Oshinski received research contract funding from Siemens Medical, the maker of the MRI scanner used in this study. The remaining authors report no conflicts of interest. Address correspondence to Ranliang Hu, MD, Department of Radiology and Imaging Sciences, Emory University School of Medicine, 1364 Clifton Road NE, Suite BG20, Atlanta, GA 30322; E-mail: [e-mail] Hu et al: J Neuro-Ophthalmol 2019; 39: 35-40 and significantly reduced CP (median 18, IQR 16-19 cm H2O, P , 0.001) after CSF removal (IQR 13-16 mL). Within patients, mean ONAs (sagittal and axial) were significantly lower before (162 ± 9°, 163 ± 10°) than after (168 ± 7°, 169 ± 5°) LP (P = 0.001, 0.008, respectively). Interrater reliability was higher with sagittal ONA measurements (0.89) than axial (0.72). Conclusions: ONA changes with short-term CSF pressure reduction in patients with IIH, establishing optic nerve tortuosity as a dynamic process related to CSF status. Journal of Neuro-Ophthalmology 2019;39:35-40 doi: 10.1097/WNO.0000000000000643 © 2018 by North American Neuro-Ophthalmology Society I diopathic intracranial hypertension (IIH) is a disorder of elevated intracranial pressure (ICP) without a known cause (1,2). Neuroimaging, particularly MRI, has emerged not only as an important tool to exclude intracranial pathologies such as mass lesion or cerebral venous sinus thrombosis, but also as supportive evidence of IIH (3,4). MRI findings including optic nerve tortuosity, distension of the optic nerve sheath, flattening of the posterior sclera, optic nerve head protrusion, and "empty" sella turcica have been described in the setting of IIH (5-9). Conversely, straightening of the optic nerve and decreased cerebrospinal fluid (CSF) within the optic nerve sheath has been reported to be present in intracranial hypotension (10). Therefore, the degree of optic nerve tortuosity and optic nerve sheath diameter may represent a dynamic process that responds to changes in ICP. We sought to explore this relationship by measuring the tortuosity and diameter of the optic nerve on high-resolution MRI before and after lumbar puncture (LP) with CSF removal and pressure reduction. We hypothesized that optic nerve sheath tortuosity and diameter would correlate with CSF pressure status and may provide useful information in the assessment of pressure status in patients 35 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution with IIH. The cranial-caudal position of the diencephalon and brainstem also was measured as a surrogate for shift in brain position before and after CSF removal. MATERIALS AND METHODS Patients This prospective study was approved by our institutional review board, and written informed consent was obtained from all patients. Patients presenting to our institution between September 2014 and September 2016 with a suspected diagnosis of IIH and referred for brain MRI and diagnostic LP were enrolled. The diagnosis of suspected IIH was made by 1 of 3 board-certified neuro-ophthalmologists based on signs and symptoms of elevated ICP such as headache, transient visual obscurations, and papilledema. Patients were excluded if they had contraindications to MRI. Exclusion criteria also included presence of an intracranial mass, hydrocephalus, or venous sinus thrombosis on contrast enhanced MRI/magnetic resonance venography. Imaging and Lumbar Puncture Protocol All patients and healthy controls underwent MRI at 3T (Tim Trio; Siemens, Erlangen, Germany) in the supine position. Patients underwent the following protocol consisting of brain MRI and fluoroscopically guided LP by 1 of 2 board-certified neuroradiologists: 1. MRI including sagittal T2 Sampling Perfection with Application optimized Contrasts using different flip angle Evolution (SPACE). 2. The patient was then taken immediately to the fluoroscopy suite and an LP was performed in the prone position with a 22-gauge spinal needle and opening pressure (OP) measured using CSF manometry. 3. CSF was withdrawn and the closing pressure (CP) measured. 4. The patient was then taken immediately back to MRI for imaging including a repeated sagittal T2 SPACE. The sagittal SPACE sequence used the following parameters for all patients: time to recovery 1,500 ms, time to echo 226 ms, flip angle 100, field of view 22.5 cm, matrix 256 · 256, and slice thickness 1.5 mm. Median time from the first sagittal SPACE to OP measurement was 61.5 minutes (interquartile range [IQR] 48.8-81.8 minutes). Median time from the CP measurement to the post-LP sagittal SPACE was 30.0 minutes (IQR 25.3-31.8 minutes). Image Analysis Image analysis was performed independently by 2 boardcertified neuroradiologists on anonymized and randomized DICOM images. They were aware that all patients were suspected of having IIH but were blinded to the demographics, CSF pressure, and LP status of each set. Measurements were performed on multiplanar T2 SPACE images using a DICOM viewer (Osirix; Pixmeo, Bernex, Switzerland). Oblique sagittal and axial images were rendered using multiplanar reformat tool to depict the optic nerve in its longest continuous profile. The angle of the greatest bend in the optic nerve in the sagittal and axial planes was measured for each eye for each examination (Fig. 1). The angle formed by long axis of the orbit and long axis of the globe also was measured to assess for the direction of horizontal FIG. 1. T2 MRI: Sagittal views show a decrease in optic nerve angle (ONA) before (A) and after (B) lumbar puncture. Axial images demonstrate a reduction of optic nerve diameter and ONA before (C) and after (D) lumbar puncture. 36 Hu et al: J Neuro-Ophthalmol 2019; 39: 35-40 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution line), and relative cranial-caudal position of the diencephalon (the distance of the mammillary body from the most cranial point of the dorsum sella) (Fig. 2B). Statistical Analysis Statistical analysis was performed using Stata (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP), with significance defined as P , 0.05. Continuous variables were reported as mean and standard deviation or median and IQR. Discrete variables were reported as counts (n) and percentages (%). Mean measurements for the 2 readers were used for further statistical analysis when applicable. Correlation between continuous variables was performed with Pearson correlation coefficient. Interobserver agreement was evaluated with interclass correlation coefficient. Measurements before and after LP were compared using paired t test. Multivariate linear regression was performed with pressure as dependent variable, and optic nerve angles (ONAs), optic nerve diameters, obex, and mammillary body position as independent variables. RESULTS A total of 10 patients with clinical diagnosis of IIH were enrolled in the study; all were female; 6 were black and 4 were white, with median age 29 years (IQR 25-32 years). Median body mass index was 35.8 kg/m2 (IQR 29.1-41.2 kg/m2). The individual CSF pressure measurements before and after LP, as well as average sagittal and axial ONA measurements are shown in Table 1. Most patients experienced an increase in sagittal ONA (9/10) and axial ONA (8/10) after LP consistent with relative optic nerve straightening. The patients who did not have an increased ONA in 1 plane had increase in the other plane. As a result, all patients had an increase in average ONA after LP (Fig. 3). All patients had elevated OP (.25 cm H2O) with median OP 37 cm H2O (IQR 34-41 cm H2O). The median volume of CSF removed FIG. 2. T2 MRI. A. Gaze angles are measured between the midline plane of the globe and horizontal line bisecting the orbit. B. Mammillary body position is measured from inferior aspect of the mammillary body to the line connecting the dorsum sella and anterior point of the tentorium. The obex position is measured from the obex to the basion-opisithion line. and vertical gaze (Fig. 2A). Additional variables measured were optic nerve sheath diameter (2 mm posterior to the optic nerve insertion), relative cranial-caudal position of the brainstem (distance of the obex from the basion-opisthion TABLE 1. Age of patients and measurement of optic nerve angle on MRI pre- and post-lumbar puncture Pre-LP MRI Lumbar Puncture Post-LP MRI Patient Age Sag ONA Ax ONA OP CP CSF Sag ONA Ax ONA 1 2 3 4 5 6 7 8 9 10 32 32 25 19 28 42 25 18 28 30 159 168 161 148 170 149 173 161 164 172 158 164 167 139 173 159 168 158 170 173 46 42 33 26 37 36 41 40 35 32 19 22 16 15 20 17 18 16 15 19 22 13 15 11 7 20 16 15 15 14 166 166 168 159 172 157 176 171 172 178 165 172 171 159 173 168 167 170 174 176 Ax ONA, average axial optic nerve angle between both eyes in degrees; CP, closing pressure in cm of water; CSF, volume cerebrospinal fluid removed in milliliters; LP, lumbar puncture; OP, opening pressure in cm of water; Sag ONA, average sagittal optic nerve angle between both eyes in degrees. Hu et al: J Neuro-Ophthalmol 2019; 39: 35-40 37 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 3. Average optic nerve angle (left and right eyes; sagittal and axial views) in our 10 patients' pre- and post-lumbar puncture. LP, lumbar puncture. to drop CSF pressure to within normal limits (,25 cm H2O) was 15 mL (IQR 13-16 mL). CP was significantly less than OP (median 18 cm H2O, IQR 16-19 cm H2O, P , 0.001) after removal of CSF. The ONA as measured before and after LP is included in Table 2, along with the vertical gaze angle, optic nerve diameter, distance of the mammillary body from the dorsum sella to the line connecting the dorsum sella and anterior tentorium, and distance of the obex from the foramen TABLE 2. Optic nerve and anatomical measurements before and after lumbar puncture Pre-LP Sag ONA Right Left Average Axial ONA Right Left Average Vertical gaze Right Left ON diameter Right Left Obex Mam body Post-LP Mean SD Mean SD P value 163 164 162 8.6 9.5 8.8 168 168 168 7.2 6.8 6.7 0.002 0.002 0.001 161 165 163 11.4 9.1 10.1 168 171 169 6.8 3.8 5.1 0.007 0.017 0.008 174 173 5.3 6.5 171 170 7.9 7.5 0.250 0.057 0.8 0.5 1.7 1.7 0.369 0.027 0.079 0.919 7.3 7.1 6.4 1.9 1.0 0.8 1.8 1.7 7.2 6.6 5.7 1.8 Bold entries denote values that are statistically significant with P , 0.05. Ax ONA, axial optic nerve in degrees; LP, lumbar puncture; Mam body, mammillary body distance to the dorsum sella in mm; obex, distance to the foramen magnum in mm; ON diameter, optic nerve sheath diameter in mm; Sag ONA, sagittal optic nerve angle in degrees; vertical gaze angle in degrees. 38 magnum. There was a significant increase in ONA both in the sagittal and axial planes after LP and CSF pressure reduction (P = 0.001, P = 0.008, respectively). The average change in sagittal ONA was 6.1° (±3.8°) and axial ONA was 6.5° (±6.1°) before and after LP. There was no significant difference in gaze before and after LP either in the vertical (right, P = 0.06; left, P = 0.91) or horizontal (right, P = 0.16; left, P = 0.91) plane. The optic nerve sheath diameter was significantly reduced after LP in the left eye (7.1 vs 6.6 mm, P = 0.027) but not significantly reduced in the right eye (7.3 vs 7.2 mm, P = 0.369). There was no significant change in obex or mammillary body position before and after LP (P = 0.079 and 0.919). Linear regression showed that sagittal ONA, axial ONA, optic nerve diameter, mammillary body position, and obex position were all not significant univariate predictors of CSF pressure on the group level. There was no significant correlation between the change in ONA and pressure change (sag ONA r = 20.245; P = 0.50, axial ONA r = 20.377; P = 0.28) or with the amount of CSF withdrawn (sag ONA r = 0.203; P = 0.57, axial ONA r = 0.04; P = 0.91). The interrater agreement was excellent for sagittal and axial ONA (intraclass correlation coefficient [ICC] 0.89, 0.72, respectively) and for mammillary body position (ICC 0.80), fair for obex position (ICC 0.56), and poor for optic nerve sheath diameter (ICC 0.25). DISCUSSION To the best of our knowledge, our study is the first to quantitatively assess the tortuosity of the optic nerve in patients with IIH before and after LP. Our results demonstrate that tortuosity of the optic nerve is associated with the CSF pressure status of the patient, in that removal of CSF and reduction of CSF pressure into the normal range immediately resulted in significantly decreased angle in the axial and sagittal bend of the optic nerve. Hu et al: J Neuro-Ophthalmol 2019; 39: 35-40 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution The mechanism of this change is likely decreased CSF distension of the perioptic subarachnoid space, leading to shortening and decreased bending of the optic nerve sheath apparatus between the relatively fixed globe and optic canal. We found sagittal ONA measurements to have higher interrater reliability than axial ONA, likely attributable to the greater ease of reformatting oblique sagittal image to display the optic nerve in a continuous profile. It should be noted that there was significant intersubject variation in ONA both before and after LP, ranging from 148 to 172° before LP. This is expected given earlier studies reporting subjective vertical tortuosity is only present in approximately 40% of patient with IIH (11-13). Innate anatomical variation, as well as possibly severity and chronicity of IIH, is likely to contribute to the intersubject variation and limit the utility of subjective vertical tortuosity as a primary diagnostic tool. Our study did not find a significant relationship between ONA and CSF pressure between patients. Variable delays between OP/CP measurement and MRI between patients also may contribute to variability in the ONA relationship with pressure. CSF volume would begin to replace soon after LP, so the postLP MRI images may be acquired with a higher OP than the measured CP. Alternatively, it may take some time for tortuosity to decrease because of chronicity or initial extent of pressure elevation. However, there was a consistent increase in ONA on an individual level after CSF removal, confirming our hypothesis that optic nerve tortuosity is a dynamic process and serial quantitative measurement may be useful to noninvasively monitor individual patient CSF pressure status. Although the results of our study show that changes in ONA with CSF pressure reduction may be seen in patients with IIH, it is also true that the ONA itself did not correlate with OP measurements. It is possible that the change in angle may simply indicate a finding that is easily discernable on neuroimaging and also may occur in those who have tortuous optic nerves without IIH. Interestingly, the optic nerve diameter was not consistently decreased in our sample population by CSF removal and was only statistically significant in the left eye. This may be explained by small magnitude of change (,1 mm), which is likely pushing the limits of the precision of our measurement technique, and our relative small sample size. A sonographic study of optic nerve sheath diameter in patients with IIH found that optic nerve sheath diameter was significantly reduced from 6.4 ± 0.6 mm (both eyes) to 5.8 ± 0.7 mm (right) and 5.9 ± 0.7 mm (left) after LP (14). The mean optic nerve sheath diameters are greater in our study, which could be due to difference in technique (MRI vs ultrasound), location of measurement, or different severity of IIH in our patient populations. The position of the mammillary body relative to the dorsum sella and of the obex relative to the foramen magnum was included in our study as surrogates for Hu et al: J Neuro-Ophthalmol 2019; 39: 35-40 position of the diencephalon and the brainstem. Sagging of the midline brain structures can be observed in the context of intracranial hypotension, but we did not find significant downward displacement of these structures with reduction of ICP. This is probably due to the fact that the pressure was reduced to a normal, not hypotensive, range, or that brain sagging may represent a chronic rather than acute process. The limitations of our study include small sample size and lack of normal control population. However, our experimental design measuring each subject before and after intervention allows each patient to serve as their own control. This allowed us to find significant intrasubject differences, despite the small sample size. In addition, patient gaze may have an impact on the tortuosity of the optic nerve. We accounted for this by measuring the horizontal and vertical gaze of the patients, which did not differ significantly between imaging sessions. However, future studies could control for this variable by having the subject fixate on a predetermined position. Finally, we did not have long-term follow-up on our patients to determine whether the ONA angle returned to pre-LP levels after time. Because LP only results in transient reduction in CSF pressure, we would expect optic nerve tortuosity to return with rising CSF pressure over time, unless treatment or intervention is performed. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: Nancy J. Newman, Valerie Biousse, Beau B. Bruce, and Amit M. Saindane; b. Acquisition of data: John Holbrook, Deqiang Qiu, and John Oshinski; c. Analysis and interpretation of data: Ranliang Hu and Amit M. Saindane. Category 2: a. Drafting the manuscript: Ranliang Hu and Amit M. Saindane; b. Revising it for intellectual content: Ranliang Hu, John Holbrook, Nancy J. Newman, Valerie Biousse, Beau B. Bruce, Deqiang Qiu, John Oshinski, and Amit M. Saindane. Category 3: a. Final approval of the completed manuscript: Ranliang Hu, John Holbrook, Nancy J. Newman, Valerie Biousse, Beau B. Bruce, Deqiang Qiu, John Oshinski, and Amit M. Saindane. REFERENCES 1. Binder DK, Horton JC, Lawton MT, McDermott MW. Idiopathic intracranial hypertension. Neurosurgery. 2004;54:538-551. 2. Ball AK, Clarke CE. Idiopathic intracranial hypertension. Lancet Neurol. 2006;5:433-442. 3. Friedman DI, Liu GT, Digre KB. Revised diagnostic criteria for the pseudotumor cerebri syndrome in adults and children. Neurology. 2013;8:1159-1165. 4. Mallery R, Woo J, Tamhankar M, Shindler K, Chen Y, Medina E, Digre K, Friedman D, Liu G. MRI Findings of elevated intracranial pressure in pseudotumor cerebri syndrome with and without papilledema. Neurology. 2014;82(suppl 10):S38.006. 5. Bidot S, Saindane AM, Peragallo JG, Bruce BB, Newman NJ, Biousse V. Brain imaging in idiopathic intracranial hypertension. J Neuroophthalmol. 2015;35:400-411. 6. Holbrook J, Saindane AM. Imaging of intracranial pressure disorders. Neurosurgery. 2017;80:341-354. 39 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 7. Suzuki H, Takanashi J, Kobayashi K, Nagasawa K, Tashima K, Kohno Y. MR imaging of idiopathic intracranial hypertension. AJNR Am J Neuroradiol. 2001;22:196-199. 8. Yuh WT, Zhu M, Taoka T, Quets JP, Maley JE, Muhonen MG, Schuster ME, Kardon RH. MR imaging of pituitary morphology in idiopathic intracranial hypertension. J Magn Reson Imaging. 2000;12:808-813. 9. Hartman AJ, Soares BP, Bruce BB, Saindane AM, Newman NJ, Biouse V, Peragallo JH. Imaging features of idiopathic intracranial hypertension in children. J Child Neurol. 2017;32:120-126. 10. Holbrook JF, Hudgins PA, Bruce BB, Saindane AM. Novel orbital findings of intracranial hypotension. Clin Imaging. 2017;41:125-131. 40 11. Agid R, Farb RI, Willinsky RA, Mikulis DJ, Tomlinson G. Idiopathic intracranial hypertension: the validity of crosssectional neuroimaging signs. Neuroradiology. 2006;48:521-527. 12. Maralani PJ, Hassanlou M, Torres C, Chakraborty S, Kingstone M, Patel V, Zackon D, Bussière M. Accuracy of brain imaging in the diagnosis of idiopathic intracranial hypertension. Clin Radiol. 2012;67:656-663. 13. Brodsky M. Magnetic resonance imaging in pseudotumor cerebri. Ophthalmology. 1998;105:1686-1693. 14. Bäuerle J, Nedelmann M. Sonographic assessment of the optic nerve sheath in idiopathic intracranial hypertension. J Neurol Sci. 2016;361:122-127. Hu et al: J Neuro-Ophthalmol 2019; 39: 35-40 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.