Accuracy of brain imaging in the diagnosis of idiopathic intracranial hypertension

Abstract

To investigate the accuracy of individual and combinations of signs on brain magnetic resonance imaging (MRI) and magnetic resonance venography (MRV) in the diagnosis of idiopathic intracranial hypertension (IIH). This study was approved by the institutional research ethics board without informed consent. Forty-three patients and 43 control subjects were retrospectively identified. Each patient and control had undergone brain MRI and MRV. Images were anonymized and reviewed by three neuroradiologists, blinded to clinical data, for the presence or absence of findings associated with IIH. The severity of stenosis in each transverse sinus was graded and summed to generate a combined stenosis score (CSS). The sensitivity, specificity, and likelihood ratios (LR) were calculated for individual and combinations of signs. Partially empty sella (specificity 95.3%, p<0.0001), flattening of the posterior globes (specificity 100%, p<0.0001), and CSS <4 (specificity 100%, p<0.0001) were highly specific for IIH. The presence of one sign, or any combination, significantly increased the odds of a diagnosis of IIH (LR+ 18.5 to 46, p<0.0001). Their absence, however, did not rule out IIH. Brain MRI with venography significantly increased the diagnostic certainty for IIH if there was no evidence of a mass, hydrocephalus, or sinus thrombosis and one of the following signs was present: flattening of the posterior globes, partially empty sella, CSS <4. However, absence of these signs did not exclude a diagnosis of IIH.