The diagnosis of idiopathic intracranial hypertension (IIH) is based primarily on the clinical findings and exclusion of secondary causes of intracranial hypertension (1). MRI is essential for the exclusion of an intracranial mass, obstructive hydrocephalus, and sinus venous thrombosis (2). Increased intracranial pressure in IIH patients may cause a number of abnormalities detected on MRI, including an empty sella, posterior globe flattening, perioptic subarachnoid space (PSAS) distension, optic nerve sheath thickening, optic nerve tortuosity, Meckel cave enlargement (2), and transverse sinus stenosis (2–7). These MRI findings alone are not diagnostic for IIH (2); however, a partially empty sella, flattening of the posterior globe, and bilateral transverse sinus stenosis seem to be the MRI findings most sensitive for IIH (8).
Although the neuroimaging findings indicative of IIH have been well established, there are few reports investigating the reversibility of such features (9,10). Our study aimed to discern the extent to which such MRI findings are reversible with treatment.
This retrospective, observational study surveyed IIH patients referred to the neurology clinic at Gazi University Hospital, Ankara, Turkey, between March 2013 and December 2017. Patients were diagnosed with IIH according to the modified Dandy criteria (11,12). Of the 75 patients who were diagnosed with IIH, IIH-associated symptoms and papilledema had resolved in 25; 10 of the 25 patients underwent pre-treatment and post-treatment brain MRI scans for evaluation. It was these 10 patients who were included in our study. All were treated with carbonic anhydrase inhibitors including acetazolamide and topiramate. None was treated with a lumboperitoneal shunt or optic nerve sheath fenestration, and neuro-ophthalmologic examinations of the patients ultimately became normal. After cessation of medical therapy, some patients continued to experience headaches, necessitating a post-treatment brain MRI. Demographics and clinical data collected were as follows: age, gender, height, weight, body mass index (BMI), papilledema grade according to the modified Frisen scale on the most affected eye, lumbar puncture (LP) opening pressure (LPOP), and the time between the resolution of symptoms and post-treatment MRI. The age- and gender-matched control group consisted of 10 patients who had normal brain imaging without any neurological findings. The study was approved by the Gazi University Institutional Review Board (No: 2017-434).
Patients and controls underwent 1.5T or 3T brain MRI scans (Magnetom Aera E11 and Magnetom Verio syngo MR B17; Siemens, Erlangen, Germany) using a standard head coil according to a standardized protocol. Axial and sagittal T1 (repetition time [TR]: 400–650 ms and echo time [TE]: 8–15 ms), axial and coronal T2 (TR: 4,000–5500 ms and TE: 85–125 ms), and axial fluid-attenuated inversion recovery (FLAIR) (TR: 8,500–9000 ms, inversion time [TI]: 2,435–2500 ms, and TE: 105–110 ms) sequences were taken. The remaining imaging parameters were as follows: field of view (FOV), 24 cm; 256 × 256 matrix; 3-mm slice thickness; and 1-mm slice gap. Diffusion-weighted (DW) images (TR: 15,000; TE, 90 ms; slice thickness, 5 mm; intersection gap, 0; 128 × 128 matrix; and FOV, 23 cm) were obtained using a spin-echo echo-planar imaging sequence with b of 0 and 1,000 s/mm2.
After the routine evaluation of the brain MRI scans, 2 experienced neuroradiologists (M.U. and N.T.), who were blinded to the patients' clinical data, extensively assessed the neuroimaging findings for IIH. Sellar configurations were evaluated in midsagittal T1 images and graded as normal, partial empty sella, or empty sella. The height of the midsagittal pituitary gland was measured in T1 or T2 sagittal images. Optic nerve thickness (ONT) and optic nerve sheath thickness (ONST) were measured using coronal T2 images in the intraorbital portion of the optic nerve just behind the globe. PSAS was calculated by subtracting ONT from ONST. Using axial FLAIR images, globe configuration was graded as 1, indicating normal convexity; 2, indicating flattened sclera; or 3, indicating optic nerve head (ONH) protrusion (concave globe). High signal intensity in the ONH was assessed through DW imaging (DWI). Horizontal tortuosity of the optic nerve is shown on axial plane T2 images. These MRI findings are shown in Figure 1.
Patients' MRI findings were recorded twice: once with papilledema and again after resolution of papilledema. Neuroradiologists graded the MRI findings based on imaging grading scales used in similar investigations (Table 1) (7,9).
All data were analyzed using SPSS (Chicago, IL, version 21). Data normality was tested using the Shapiro–Wilk test. Demographic data were expressed as mean (±SD). Because the data were not normally distributed, comparisons were performed using the Mann–Whitney U test or Wilcoxon signed-rank test and were expressed as a median (interquartile range [IQR]). Associations between non-normally distributed and ordinal variables, the correlation coefficients, and their corresponding levels of significance were calculated using the Spearman test. Analysis of variance tests were conducted followed by Bonferroni post hoc comparisons tests. Values of P < 0.05 were considered statistically significant.
This study included 10 female patients with a median age at IIH diagnosis of 29 years (IQR: 22.75/35.5). The pre-treatment BMI median value was 33.57 kg/m2 (IQR: 30.18/41.27), whereas the post-treatment BMI median value was 28.1 kg/m2 (IQR: 23.8/29.7). All the patients were (100%) overweight (BMI ≥25 kg/m2). LP was performed within 2 days of the fundus examination and MRI studies; one patient's LPOP was 21 cmH2O. This patient was diagnosed with probable IIH (12). Demographic and clinical data of the patients are summarized in Table 2.
Patients with papilledema exhibited partial empty or empty sella (7/10), posterior globe flattening or protrusion (9/10), horizontal optic nerve tortuosity (8/10), or ONH (4/10). After resolution of the papilledema, all of the patients showed improvement in 2 or more of these MRI findings (Table 3). The midsagittal pituitary gland height, ONST, and PSAS were significantly different in all pairwise comparisons of the groups. For the IIH patients, the height of the midsagittal pituitary gland was lower in the pre-treatment group than the post-treatment group. After treatment, the height of the midsagittal pituitary gland was still lower in the IIH patients than in the controls (Table 3). Measurements of the ONST and PSAS in the IIH patients were higher in both pre-treatment and post-treatment groups relative to controls (See Supplemental Digital Content 1, Fig. E1, https://links.lww.com/WNO/A368). The difference in sellar configuration, globe configuration, and horizontal tortuosity between the IIH pre-treatment group and controls became insignificant after treatment. We found no difference among the 3 groups in terms of optic nerve hyperintensity or thickness. Age, papilledema Frisen grade, duration of disease, and LPOP were not significantly associated with sellar configuration or height of the pituitary gland. Significant, strong, positive correlations were noted between BMI and ONST (r = 0.83) (P = 0.006) as well as between BMI onset and PSAS (r = 0.762) (P = 0.017) in the IIH pre-treatment group; as BMI increased, changes in ONST and PSAS also increased. Furthermore, we found a strong negative correlation between BMI and height of the midsagittal pituitary gland (r = −0.73; P < 0.0001) (See Supplemental Digital Content 2, Fig. E2, https://links.lww.com/WNO/A369).
In our study, we demonstrated the improvement in midsagittal pituitary gland height, ONST, and PSAS after resolution of papilledema. However, midsagittal pituitary gland height was still lower, and the measurements of ONST and PSAS were still higher in the IIH post-treatment group than in controls. In addition, we found sellar configuration, globe configuration, and optic nerve horizontal tortuosity varied between the IIH pre-treatment group and controls, but the difference became insignificant after treatment. ONH hyperintensity and ONT showed no difference between all pairs of groups.
Sellar Configuration and Pituitary Gland Height
Although empty sella is a nonspecific finding in intracranial hypertension, it reflects the chronicity of increased intracranial pressure and facilitates the diagnosis of IIH (4,13). Morphometric MRI studies have revealed that the reduced height of the pituitary gland is associated more with the enlargement of the bony sella than with the reduction of the actual pituitary gland size (14). Yuh et al (4) hypothesized that cumulative cerebrospinal fluid (CSF) pressure causes sellar reconfiguration, beginning in the subclinical phases of IIH. Only a few studies and case reports have studied whether remodeling of the pituitary gland and sella turcica is reversible (9,10,15,16). Ranganathan et al (10) found that the area and height of the pituitary gland increase in IIH patients after treatment. Their findings suggest that the pituitary gland is not compressed, but rather deformed, and that the filling of the suprasellar cistern induces an empty sella. We observed that empty sella was more common in the pre-treatment group of IIH patients than in healthy controls. The frequency of empty sella was similar in the pre-treatment and post-treatment patient groups. It was also similar within the post-treatment group and controls. However, height of the pituitary gland of the IIH pre-treatment and post-treatment groups was significantly lower relative to that of healthy controls. The pituitary gland height of IIH patients increased after months of treatment but was still lower than that of the control group. The filling of the subarachnoid cisterns with CSF can thus be reduced, but the expanded spaces and bony changes do not fully recover. This same explanation could also be adopted for data collected from healthy controls because anatomical defects of the diaphragm sella is a common variation among the normal population (17,18).
Normal convexity of the posterior globe reflects the equilibrium between intracranial and intraocular pressure (2,5). As observed on axial FLAIR images, the flattening or concavity of the posterior sclera indicates an increase in CSF pressure through the PSAS (5). Previous studies reported this MRI finding to be highly specific of, but not sensitive to, IIH (14,19). Yet, Agid et al (19) found that flattening of the sclera is the only sign that conclusively identifies IIH. Our study confirmed the high specificity of the globe configuration both in the pre-treatment and post-treatment groups. We further demonstrated that the IIH-induced altered globe configuration improves after treatment.
Optic Nerve Sheath Thickness
IIH is characterized by an increase in intracranial pressure, which causes distension of the PSAS and enlargement of the ONS (2). Previous studies in adults and children showed that the measurement of ONST through neuroimaging assists in the diagnosis of IIH (20). Morphometric and volumetric MRI analyses have proven useful especially in formulating cutoff values for accurate diagnosis (3,21). We demonstrated similar findings after the treatment of IIH: the average diameter of ONS was found to be reduced in IIH patients but remained higher than that in healthy controls. Considering the elasticity of the ONS, the post-treatment ONS diameter may become normalized over the course of time.
Optic Nerve Head Hyperintensity
ONH hyperintensity assessed with DWI is a reliable, highly specific indicator of papilledema (22,23). However, Salvay et al (23) found no association between lower grades of papilledema and ONH hyperintensity. The absence of ONH hyperintensity thus does not exclude papilledema. In our study, the lower grade of papilledema among the IIH patients may account for the lack of a significant difference in ONH hyperintensity between the groups. Viets et al (22) posited that the pathophysiology of axoplasmic stasis in papilledema may be better explained by an ischemic mechanism causing compression of ciliary circulation rather than a mechanical compression that leads to the axoplasmic buildup; only severe papilledema and axoplasmic stasis would, therefore, cause structural damage and consequent ONH hyperintensity.
Horizontal Tortuosity of the Optic Nerve
Kinking of the optic nerve, as observed in the sagittal and axial planes, defines as tortuosity of the optic nerve. Horizontal or vertical tortuosity is related to distension of the optic nerve sheath between its proximal and distal fixation points (2,6). Although the sensitivity of this MRI finding is not high, horizontal tortuosity is reportedly more specific to IIH than vertical tortuosity (5,24). Görkem et al (24) found that horizontal tortuosity has 68% sensitivity and 83% specificity in pediatric IIH patients. Our findings in adults were similar: the frequency of horizontal tortuosity was significantly higher in the pre-treatment IIH group than in controls, whereas there was no difference between the post-treatment IIH group and controls. This finding implies that horizontal tortuosity is indicative of IIH in the pre-treatment group but is not a valuable follow-up indicator of IIH.
Two previous reports examined the reversibility of MRI findings attributed to IIH (9,10). Ranganathan et al (10) evaluated the pituitary gland height and area in IIH after 1 week of lumbar punction and initiation of medical therapy. They determined that the relative increase in the area of the pituitary gland was more sensitive than the height of the pituitary gland to assess improvement (10). In our study, there was a relative increase in the pituitary gland height after treatment, but the difference between post-treatment and control groups remained statistically significant. Chang et al (9) investigated the persistence of MRI findings in patients with IIH by comparing the data of resolved papilledema and active papilledema with a control group. They were unable to show a significant difference in sellar and globe configurations among the 3 groups. By contrast, we demonstrated that 2 or more of the abnormal MRI findings improved in all patients after treatment. However, we found that enlargement of CSF-filled SASs recovered only partially.
Morphometric MRI finding characteristics to IIH are valuable not only for diagnosis of the disease but also in assessing recovery during follow-up. This study demonstrated that several of these MRI findings are reversible to a certain extent after treatment: the ONST is decreased, and the compressed pituitary gland shows re-expansion, but the sellar configuration, optic nerve hyperintensity on DWI, and horizontal tortuosity do not differ from those of controls. However, enlarged SASs filled with CSF appear to be reduced, whereas the ONST and pituitary gland height are not fully normalized in the post-treatment IIH group. Limited reversal of the MRI findings in IIH cases after treatment may also be associated with the time-dependent elasticity of pressure-naive anatomical structures and possible bony erosions.
This study was limited by its small sample size and a control group that was not matched for BMI. Future research should include more IIH patients with adequately matched healthy controls. In addition, the data in our retrospective study varied among patients in the time between the resolution of papilledema and performance of the post-treatment MRI. Finally, we did not include a visual field evaluation in our study.
The authors thank Editage (www.editage.com) for English language editing. The authors also thank Prof. Dr Ihsan Alp for careful review of statistical analysis.
1. Wall M. Idiopathic intracranial hypertension. Neurol Clin. 2010;28:593–617.
2. Bidot S, Saindane AM, Peragallo JH, Bruce BB, Newman NJ, Biousse V. Brain imaging in idiopathic intracranial hypertension. J Neuroophthalmol. 2015;35:400–411.
3. Hoffmann J, Huppertz HJ, Schmidt C, Kunte H, Harms L, Klingebiel R, Wiener E. Morphometric and volumetric MRI changes in idiopathic intracranial hypertension. Cephalalgia. 2013;33:1075–1084.
4. 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.
5. Degnan AJ, Levy LM. Pseudotumor cerebri: brief review of clinical syndrome and imaging findings. AJNR Am J Neuroradiol. 2011;32:1986–1993.
6. Passi N, Degnan AJ, Levy LM. MR imaging of papilledema and visual pathways: effects of increased intracranial pressure and pathophysiologic mechanisms. AJNR Am J Neuroradiol. 2013;34:919–924.
7. 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.
8. Maralani PJ, Hassanlou M, Torres C, Chakraborty S, Kingstone M, Patel V, Zackon D, Bussiere M. Accuracy of brain imaging in the diagnosis of idiopathic intracranial hypertension. Clin Radiol. 2012;67:656–663.
9. Chang RO, Marshall BK, Yahyavi N, Sharma A, Huecker J, Gordon MO, McClelland C, Van Stavern GP. Neuroimaging features of idiopathic intracranial hypertension persist after resolution of papilloedema. Neuroophthalmology. 2016;40:165–170.
10. Ranganathan S, Lee SH, Checkver A, Sklar E, Lam BL, Danton GH, Alperin N. Magnetic resonance imaging finding of empty sella in obesity related idiopathic intracranial hypertension is associated with enlarged sella turcica. Neuroradiology. 2013;55:955–961.
11. Friedman DI, Jacobson DM. Diagnostic criteria for idiopathic intracranial hypertension. Neurology. 2002;59:1492–1495.
12. Friedman DI, Liu GT, Digre KB. Revised diagnostic criteria for the pseudotumor cerebri syndrome in adults and children. Neurology. 2013;81:1159–1165.
13. Brodsky MC, Vaphiades M. Magnetic resonance imaging in pseudotumor cerebri. Ophthalmology. 1998;105:1686–1693.
14. Kyung SE, Botelho JV, Horton JC. Enlargement of the sella turcica in pseudotumor cerebri. J Neurosurg. 2014;120:538–542.
15. Zagardo MT, Cail WS, Kelman SE, Rothman MI. Reversible empty sella in idiopathic intracranial hypertension: an indicator of successful therapy? AJNR Am J Neuroradiol. 1996;17:1953–1956.
16. 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.
17. Evans RW. Incidental findings and normal anatomical variants on MRI of the brain in adults for primary headaches. Headache. 2017;57:780–791.
18. Guitelman M, Garcia Basavilbaso N, Vitale M, Chervin A, Katz D, Miragaya K, Herrera J, Cornalo D, Servidio M, Boero L, Manavela M, Danilowicz K, Alfieri A, Stalldecker G, Glerean M, Fainstein Day P, Ballarino C, Mallea Gil MS, Rogozinski A. Primary empty sella (PES): a review of 175 cases. Pituitary. 2013;16:270–274.
19. 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.
20. Hirfanoglu T, Aydin K, Serdaroglu A, Havali C. Novel Magnetic resonance imaging findings in children with intracranial hypertension. Pediatr Neurol. 2015;53:151–156.
21. Hoffmann J, Schmidt C, Kunte H, Klingebiel R, Harms L, Huppertz HJ, Ludemann L, Wiener E. Volumetric assessment of optic nerve sheath and hypophysis in idiopathic intracranial hypertension. AJNR Am J Neuroradiol 2014;35:513–518.
22. Viets R, Parsons M, Van Stavern G, Hildebolt 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.
23. Salvay DM, Padhye LV, Huecker JB, Gordon MO, Viets R, Sharma A, Van Stavern GP. Correlation between papilledema grade and diffusion-weighted magnetic resonance imaging in idiopathic intracranial hypertension. J Neuroophthalmol. 2014;34:331–335.
24. Görkem SB, Doğanay S, Canpolat M, Koc G, Dogan MS, Per H, Coşkun A. MR imaging findings in children with pseudotumor cerebri and comparison with healthy controls. Childs Nerv Syst. 2015;31:373–380.