The contribution of neuroimaging to diagnosis in idiopathic intracranial hypertension

Abstract

Objective: Due to difficulties in diagnosis, the presence of patients with idiopathic intracranial hypertension (IIH) without papilledema has led to the addition of neuro-radiologic features, which are indicative of increased intracranial pressure, to new diagnostic criteria. In this study, it was aimed to investigate the inter-rater agreement regarding the evaluation of neuroimaging findings of patients diagnosed as having definite IIH and to detect their possible contribution to diagnosis. Materials and Methods: Forty-one consecutive patients with definite IIH who had both brain magnetic resonance imaging (MRI) and MR-venography (MRV) examinations were included in the study. Images were evaluated by two experienced neurologists, who were blinded to the clinical presentation of the cases, in terms of empty sella, flattened posterior globe/sclera, tortuosity of optic nerve, transverse sinus stenosis/hypoplasia, and additionally non-specific white matter lesions using a standardized form. The results were evaluated using kappa (κ) analysis in terms of inter-rater agreement. Results: The mean body mass index and cerebral spinal fluid opening pressure of the patients (38 female, 3 male) were 29.2±5.48 kg and 371±149.4 mmH2O, respectively. Six patients did not have papilledema. When MRI and MRV examinations were evaluated in terms of the presence of additional evidence supporting diagnosis, there was fair agreement for MRI (κ=0.333, p=0.021), and substantial agreement for MRV (κ=0.735, p<0.0001). Regarding subheadings, there was moderate agreement for empty sella, tortuosity of the optic nerve, and non-specific white matter lesions (κ=0.417, p=0.001; κ=0.523, p=0.001; κ=0.443, p=0.002, respectively), fair agreement for flattened posterior globe/sclera and transvers venous sinus stenosis (κ=0.291, p=0.06; κ=0.389, p=0.006, respectively), and substantial agreement for transverse venous sinus hypoplasia (κ=0.801, p<0.001). Conclusion: In our study, it was found that additional neuroradiologic findings, especially those described in the new Friedman diagnostic criteria for patients without papilledema, showed fair-to-moderate agreement, even between two experienced neurologists. We believe that it is appropriate to examine the neuroradiologic images carefully for new criteria without evaluating only mass lesions, and to take clinical findings into account while evaluating patients with IIH.