Journal of Neuro-Ophthalmology, December 2014, Volume 34, Issue 4

Correlation Between Papilledema Grade and Diffusion-Weighted Magnetic Resonance Imaging in Idiopathic Intracranial Hypertension

Correlation Between Papilledema Grade and Diffusion-Weighted Magnetic Resonance Imaging in Idiopathic Intracranial Hypertension David M. Salvay, MD, PhD, Leena V. Padhye, BA, Julie B. Huecker, MS, Mae O. Gordon, PhD, Ryan Viets, MD, Aseem Sharma, MD, Gregory P. Van Stavern, MD Background: To explore the relationship between diffusion-weighted magnetic resonance imaging (DWI) hyperintensity of the optic nerve head (ONH) and papil-ledema grade in patients with idiopathic intracranial hypertension (IIH). Methods: A retrospective chart review was conducted of patients with definitively diagnosed IIH by clinical examina-tion and visual field (VF) analysis who underwent orbital magnetic resonance imaging (MRI) within 4 weeks of diagnosis. A neuroradiologist masked to the diagnosis assessed the results of DWI for each eye independently and graded the signal intensity of the ONH into none, mild, and prominent categories. DWI grading was compared with papilledema grade and visual field mean deviation (VFMD) by Spearman rank correlation analysis and t-tests. Results: Forty-two patients were included in the study. A statistically significant difference (P = 0.0195) was found between papilledema grade and patients with prominent DWI findings (n = 16; mean papilledema grade 3.75 ± 1.25) vs mild or no ONH hyperintensity (n = 26; mean pap-illedema grade 2.79 ± 1.24) at the time of initial diagnosis. DWI hyperintensity of the ONH at diagnosis was also found to be significantly correlated with the degree of papilledema at follow-up (r = 0.39, P = 0.0183) but not with VFMD. Conclusions: We found a significant correlation between the severity of papilledema and ONH hyperintensity on DWI in patients with IIH but not with VF loss or other visual parameters. These findings may offer insight into the pathophysiology of papilledema in IIH and provide a surro-gate marker for the presence and severity of papilledema. Journal of Neuro-Ophthalmology 2014;34:331-335 doi: 10.1097/WNO.0000000000000150 © 2014 by North American Neuro-Ophthalmology Society Idiopathic intracranial hypertension (IIH) is a relatively uncommon neurological disorder characterized by increased intracranial pressure (ICP) without an identifiable etiology and is an important cause of papilledema and visual loss (1-4). Although patients with severe visual loss tend to have higher grades of papilledema, many patients with profound papilledema retain normal vision. Previous studies have investigated whether factors, such as increased cerebro-spinal fluid pressure, low intraocular pressure, or low optic nerve perfusion pressure are risk factors for or predictive of visual loss, but no strong correlations have been conclusively identified (5,6). Lacking such predictive metrics, it remains unknown whether the severity of papilledema or the presence of other clinically identifiable signs at initial pre-sentation can be correlated with final visual outcome (2). Orbital imaging offers an alternative method to assess papilledema and other structural changes in the optic nerve characteristic of IIH. These features include flattening of the perioptic subarachnoid space, enhancement of the prelaminar optic nerve head (ONH), and slit-like ventricles (7-10). In Department of Ophthalmology and Visual Sciences (DMS, JBH, MOG, GPVS), Washington University, Saint Louis, Missouri; Department of Ophthalmology and Visual Sciences (LVP), Washington University, Saint Louis, Missouri; Department of Radiology (RV, AS), Mallinckrodt Institute of Radiology, Washington University School, Saint Louis, Missouri; and Department of Neurology (GPVS), Washington University, Saint Louis, Missouri. Supported by Washington University, Department of Ophthalmol-ogy and Visual Sciences, Core Grant 5 P30 EY02687; Institute for Clinical and Translational Sciences, Grant RR023496; Biostat Core, Grant U54 RR023496; NIH Core Vision, Grant P30 EY02687; and an unrestricted grant from Research to Prevent Blindness. Supported by the NIH, National Center for Research Resources (NCRR), Grant Numbers UL1 RR024992 and TL1 RR024995); and by a Dean's Fel-lowship (Washington University School of Medicine, Washington University in Saint Louis, Missouri). G. P. Van Stavern has received attorney fees related to expert tes-timony and royalty fees from UpToDate for a review article unre-lated to the topic of the current study. The remaining authors report no conflicts of interest. Address correspondence to: Gregory P. Van Stavern, MD, Depart-ment of Ophthalmology and Visual Sciences, Washington University School of Medicine, 660 South Euclid Avenue, Campus Box 8096, Saint Louis, MO 63110-1010; E-mail: [e-mail] Salvay et al: J Neuro-Ophthalmol 2014; 34: 331-335 331 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. a previous report, we showed that no correlation exists between orbital imaging findings of papilledema and visual function using standard and high-resolution T1, T2, and fluid-attenuated inversion recovery image sequences (11). In contrast, increased signal in the ONH on diffusion-weighted magnetic resonance imaging (DWI) has been re-ported as a marker of IIH (10), but the degree to which this finding correlates with clinical parameters and outcomes re-mains unknown. We sought to assess whether a correlation exists between radiological and clinical parameters associated with papilledema. METHODS Study Population A retrospective chart review was conducted of consecutive patients seen in our neuro-ophthalmology clinic from April 2009 through April 2013. Inclusion criteria included: definite or probable diagnosis of IIH using established diagnostic criteria (12,13), age at diagnosis 18 years or older, papilledema secondary to IIH, a complete neuro-ophthalmological examination (with fundus photo-graphs, papilledema grade, and visual field [VF] analysis) performed within 4 weeks of neuroimaging. Data Collection Approval for this study was obtained from the Washington University Human Research Protection Office before data accumulation, and all HIPAA regulations regarding the use of personal health information were strictly followed. The following demographic, clinical, and radiological data were collected for all patients: initial and most recent examination dates, date of birth, gender, race, height, weight, best corrected Snellen visual acuity, color vision assessment (using Ishihara color plates), visual field mean deviation (VFMD; automated perimetry was performed on all patients except one), papilledema grade, and lumbar puncture opening pressure. Papilledema grade was determined by a neuro-ophthalmologist using the modified Frisén staging scheme (14). Papilledema grade was recorded and ana-lyzed in 1 of 2 ways: either using the neuro-ophthalmol-ogist's grade from the original chart note or for those eyes containing noninteger grades, the neuro-ophthalmologist re-graded those eyes using color fundus photographs (Fig. 1). MRI was performed on 1.5T (Symphony, Esprit, or Sonata; Siemens, Erlangen, Germany) or 3T (Trio; Siemens) scanners. The parameters used for DWI included an average TR of 3,300 milliseconds, TE of 100 milliseconds, and section thickness of 5 to 7 mm with an interslice gap of 1.5 mm. An experienced neuroradiologist masked to the details of patient's clinical assessment evaluated images obtained by DWI (Fig. 1) and categorized the signal intensity of the ONH on DWI into a 3-point scale (normal-Grade 0; mild hyperintensity-Grade 1; and prominent hyperintensity- Grade 2), as has been previously described (15). As reported in a previous study (11), visual and imaging parameters were averaged between eyes based on the following rationale: 1) papilledema is generally highly symmetric, 2) ICP should be relatively equal and symmetrically distributed to both optic nerves, and 3) the effect of increased ICP on one eye should be equivalent to the effect on the fellow eye. This assumption was confirmed by high Spearman correlations between right and left eyes for DWI (r = 0.81). Similar results were found for VFMD (r = 0.82, r = 0.75) and papilledema grade (r = 0.90, r = 0.96) at both initial presentation and the most recent follow-up visit, respectively. Because of the high intercorrelation between eyes, conducting independent analyses for each eye individually was deemed unnecessary, and we used the mean values for right and left eyes in subsequent analyses. To assess intraobserver reliability, neuroimaging findings of 10 randomly selected patients were reassessed by the neuroradiologist, who remained masked to clinical data and the results of the initial review. This was performed at least 1 month after the initial interpretation. The degree of signal hyperintensity was regraded and found to be highly correlated with the initial assessment. Main Outcome Measures The major outcome analysis involved correlating the main clinical parameters of papilledema grade, visual acuity, and VFMD at presentation and follow-up clinical visit with the ONH signal intensity on DWI sequences conducted within 4 weeks of initial diagnosis. Statistical Analysis The clinical parameters of papilledema grade, VFMD, and visual acuity were correlated with MRI features. Statistical analysis using t-tests and Spearman rank correlations was performed using SAS (version 9.2; Statistical Analysis Software, Cary, NC). RESULTS Forty-two patients (41 female, 1 male) with a mean age of 31.2 ± 10.0 years were included in the study. Thirty patients (71.4%) were white, whereas 12 (28.6%) were African- American. Thirty-six of these patients (85.7%; 35 female, 1 male; mean age 31.4 ± 10.3 years) returned for follow-up (mean time between initial diagnosis and most recent examination was 652 ± 359 days). Of these, 26 (72.2%) were white and 10 (27.8%) were African-American. Demo-graphic and selected clinical parameters are summarized for all study participants at the time of diagnosis (Table 1) and at the most recent follow-up visit (Table 2). Seventeen patients had noninteger grades recorded in their chart (e.g., the original clinical note reported "Grades 3-4"), and thus an integer papilledema grade based on the 332 Salvay et al: J Neuro-Ophthalmol 2014; 34: 331-335 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. modified Frisén scale was determined after reviewing color fundus photographs. A statistically significant difference (P = 0.0195) was found between papilledema grade and patients with prom-inent ONH hyperintensity (n = 16; mean papilledema grade 3.75 ± 1.25) vs those with none/mild ONH hyper-intensity (n = 26; mean papilledema grade 2.79 ± 1.24) on DWI at the time of initial diagnosis. Similarly, a statistically significant correlation (r = 0.39, P = 0.0183) was found between DWI findings at the initial visit and papilledema grade at the most recent follow-up visit. No other statisti-cally significant associations were found between DWI re-sults at the initial visit and VFMD or visual acuity either at the time of diagnosis or at the most recent follow-up visit. DISCUSSION We sought to determine whether DWI findings at the time of diagnosis of IIH were associated with clinical parameters such as papilledema grade, VFMD, and visual acuity. We also examined whether these imaging findings might be predictive of final visual outcome. This hypothesis was based on the results of a case-control analysis that found hyper-intensity of the ONH on DWI in a significant number of patients with papilledema at our institution (15). DWI provides a noninvasive method to evaluate the rate of water movement in biologic tissues. The relative ease with which water diffuses within a tissue can provide important details regarding its architecture since a lack of movement, also known as restricted diffusion, implies stasis. This property is particularly useful when evaluating white matter tracts because they have been shown to be anisotropic (i.e., water diffuses more rapidly parallel to the direction of the internal fiber structure than perpendicular to it) (16). When the white matter tract becomes swollen, the anatomy of the fiber structure is disrupted, resulting in restricted diffusion. This finding has been noted in a variety of presentations associated with optic nerve edema, including ischemic optic neuropathy (17-19), rhinocerebral mucormycosis (20), and optic neuritis (21). One case report documented the presence of restricted diffusion in clinically proven papilledema because of increased ICP sec-ondary to posterior fossa brain tumor (22). Other than our previously published study (15), we are not aware of any other reports investigating the association between DWI and papil-ledema secondary to IIH. Our results indicate that prominent ONH hyperintensity on DWI is a predictor of papilledema grade at the time of diagnosis of IIH. A similar association also was found between ONH hyperintensity and papilledema grade at follow-up visits up to 3.6 years after initial diagnosis. The fact that no statistically significant association was found between ONH hyperintensity and papilledema grade in patients with less FIG. 1. A. Grade 3 papilledema is present. B. Corresponding axial diffusion-weighted image shows prominent signal intensity (arrows) of the optic discs. TABLE 1. Patient demographic characteristics and clinical parameters (average of both eyes) for study participants at time of initial diagnosis (n = 42) Characteristics Mean ± SD Range Age (yr) 31.2 ± 10.0 19 to 52 Body mass index (kg/m2) 37.0 ± 8.9 25.5 to 73.5 Lumbar puncture opening pressure (cm H2O) 33.2 ± 9.6 16 to 57 Papiledema grade 3.0 ± 1.3 0 to 5 VFMD (dB) 24.1 ± 5.7 231.7 to 4.4 DWI grade 1.1 ± 0.7 0 to 2 dB, decibel; DWI, diffusion-weighted imaging; VFMD, visual field mean deviation. TABLE 2. Patient characteristics and clinical parameters (average of both eyes) for study participants at most recent follow-up visit (n = 36) Characteristics Mean ± SD Range Age (yr) 33.2 ± 10.1 20 to 53 Papilledema grade 1.0 ± 1.4 0 to 5 VFMD (dB) 21.9 ± 2.9 215.8 to 1.3 Time between diagnosis and most recent visit (d) 652 ± 359 49 to 1,342 d, days; dB, decibel; DWI, diffusion-weighted imaging; VFMD, visual field mean deviation. Salvay et al: J Neuro-Ophthalmol 2014; 34: 331-335 333 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. severe papilledema suggest that the structural changes within the ONH are short lived and may be less damaging than those associated with more severe cases. Although this information does not provide additional clues regarding the underlying physiological changes that may be occurring at the ONH in papilledema, it does provide some insight into its pathophysiology. It is thought that the finding of hyperintensity reflects axoplasmic stasis secondary to elevated ICP due to either mechanical com-pression of axons (23) or ischemia of the ciliary circulation (24-26). This would support the hypothesis that prominent ONH intensity on DWI implies more severe axoplasmic stasis and portends a poor long-term visual prognosis. The lack of correlation observed between findings and visual parameters (visual acuity, VFMD) may result from multiple factors. First, although we were able to follow patients up to 3.6 years after diagnosis, the mean follow-up interval was slightly less than 1.8 years. Because axonal dropout in optic atrophy secondary to IIH may take years to occur and depends on both the degree and duration of ICP elevation, our follow-up interval may not have been long enough to detect clinically significant visual loss. Second, it may be that the anatomic and physiologic changes in the ONH detected with DWI and fundus examination are only loosely correlated with progressive axonal loss; that is, although the DWI is detecting localized axoplasmic stasis, stasis may not in and of itself be directly correlated with axon damage and death. Finally, volume-averaging effects and the relatively poor resolution inherent with DWI images make it difficult to investigate subtle imaging details. If available, possibly a correlation could be established with assessment of visual function. Many patients received standard imaging, with 3- to 5-mm slice thickness. Higher resolution imaging with ,1-mm slices might detect more prominent DWI hyperintensity and provide better correla-tion with visual outcomes. Additionally, it is possible that because imaging technology continues to improve and the resolution of such scans increases, quantitative analysis of apparent diffusion coefficient (ADC) values at the ONH might provide a better measure of diffusion characteristics as compared with the qualitative scale used in this study. Because of the small physical dimensions of the ONH, it is difficult to reliably include it in a region of interest for ADC measurements. REFERENCES 1. Van Stavern GP. Optic disc edema. Semin Neurol. 2007;27:233-243. 2. Wall M. Idiopathic intracranial hypertension. Neurol Clin. 2010;28:593-617. 3. Dhungana S, Sharrack B, Woodroofe N. Idiopathic intracranial hypertension. Acta Neurol Scand. 2010;121:71-82. 4. Randhawa S, Van Stavern GP. Idiopathic intracranial hypertension (pseudotumor cerebri). Curr Opin Ophthalmol. 2008;19:445-453. 5. Fraser C, Plant GT. The syndrome of pseudotumour cerebri and idiopathic intracranial hypertension. Curr Opin Neurol. 2011;24:12-17. 6. Pollak L, Zohar E, Glovinsky Y, Huna-Baron R. Reevaluation of presentation and course of idiopathic intracranial hypertension-a large cohort comprehensive study. Acta Neurol Scand. 2013;127:406-412. 7. Agid R, Farb RI, Willinsky RA, Mikulis DJ, Tomlinson G. Idiopathic intracranial hypertension: the validity of cross-sectional neuroimaging signs. Neuroradiology. 2006;48:521-527. 8. Brodsky MC, Vaphiades M. Magnetic resonance imaging in pseudotumor cerebri. Ophthalmology. 1998;105:1686-1693. 9. Lim MJ, Pushparajah K, Jan W, Calver D, Lin J. Magnetic resonance imaging changes in idiopathic intracranial hypertension in children. J Child Neurol. 2010;25:294-299. 10. Mashima Y, Oshitari K, Imamura Y, Momoshima S, Shiga H, Oguchi Y. High- resolution magnetic resonance imaging of the intraorbital optic nerve and subarachnoid space in patients with papilledema and optic atrophy. Arch Ophthalmol. 1996;114:1197-1203. 11. Padhye LV, Van Stavern GP, Sharma A, Viets R, Huecker JB, Gordon MO. Association between visual parameters and neuroimaging features of idiopathic intracranial hypertension. J Neurol Sci. 2013;332:80-85. 12. Friedman DI, Jacobson DM. Diagnostic criteria for idiopathic intracranial hypertension. Neurology. 2002;59:1492-1495. 13. Friedman DI, Liu GT, Digre KB. Revised diagnostic criteria for the pseudotumor cerebri syndrome in adults and children. Neurology. 2013;81:1159-1165. 14. Scott CJ, Kardon RH, Lee AG, Frisen L, Wall M. Diagnosis and grading of papilledema in patients with raised intracranial pressure using optical coherence tomography vs clinical expert assessment using a clinical staging scale. Arch Ophthalmol. 2010;128:705-711. 15. Viets R, Parsons M, Van Stavern G, Hildebult C, Sharma A. Hyperintense optic nerve heads on diffusion-weighted imaging: a potential imaging sign of papilledema. AJNR Am J Neuroradiol. 2013;34:1438-1442. 16. Pierpaoli C, Jezzard P, Basser PJ, Barnett A, Dichiro G. Diffusion tensor MR imaging of the human brain. Radiology. 1996;201:637-648. 17. Al-Shafai LS, Mikulis DJ. Diffusion MR imaging in a case of acute ischemic optic neuropathy. AJNR Am J Neuroradiol. 2006;27:255-257. 18. Chen JS, Mukherjee P, Dillon WP, Wintermark M. Restricted diffusion in bilateral optic nerves and retinas as an indicator of venous ischemia caused by cavernous sinus thrombophlebitis. AJNR Am J Neuroradiol. 2006;27:1815-1816. 19. Verma A, Jain KK, Mohan S, Phadke RV. Diffusion-weighted MR imaging in posterior ischemic optic neuropathy. AJNR Am J Neuroradiol. 2007;28:1839-1840. 20. Mathur S, Karimi A, Mafee MF. Acute optic nerve infarction demonstrated by diffusion-weighted imaging in a case of rhinocerebral mucormycosis. AJNR Am J Neuroradiol. 2007;28:489-490. 21. Spierer O, Ben Sira L, Leibovitch I, Kesler A. MRI demonstrates restricted diffusion in distal optic nerve in atypical optic neuritis. J Neuroophthalmol. 2010;30:31-33. 22. Pakzad-Vaezi K, Cochrane D, Sargent M, Singhal A. Conventional and diffusion- weighted magnetic resonance imaging findings in a pediatric patient with a posterior fossa STATEMENT OF AUTHORSHIP Design of study (D. M. Salvay, A. Sharma, G. P. Van Stavern); Conduct of the study (D. M. Salvay); Collection and management of data (D. M. Salvay, A. Sharma, R. Viets); Analysis of data (D. M. Salvay, J. B. Huecker, M. O. Gordon, R. Viets, A. Sharma, G. P. Van Stavern); Interpretation of data (D. M. Salvay, G. P. Van Stavern, A. Sharma); Preparation and review of manuscript (D. M. Salvay, G. P. Van Stavern, A. Sharma, J. B. Huecker, M. O. Gordon); Approval of manuscript (D. M. Salvay, G. P. Van Stavern). 334 Salvay et al: J Neuro-Ophthalmol 2014; 34: 331-335 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. brain tumor and papilledema. Pediatr Neurosurg. 2009;45:414-418. 23. Trobe JD. Papilledema: the vexing issues. J Neuroophthalmol. 2011;31:175-186. 24. McLeod D, Marshall J, Kohner EM. Role of axoplasmic transport in the pathophysiology of ischaemic disc swelling. Br J Ophthalmol. 1980;64:247-261. 25. Radius RL. Optic nerve fast axonal transport abnormalities in primates. Occurrence after short posterior ciliary artery occlusion. Arch Ophthalmol. 1980;98:2018-2022. 26. Corbett JJ, Savino PJ, Thompson HS, Kansu T, Schatz NJ, Orr LS, Hopson D. Visual loss in pseudotumor cerebri. Follow-up of 57 patients from five to 41 years and a profile of 14 patients with permanent severe visual loss. Arch Neurol. 1982;39:461-474. Salvay et al: J Neuro-Ophthalmol 2014; 34: 331-335 335 Original Contribution Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.