A 13-year-old previously healthy female presented at our institution following an acute episode of altered mental status characterized by impairment of speech, urinary incontinence, and emesis. Upon arrival, the patient described 2 prior migraine-like headaches with bilateral pulsatile tinnitus in the 2 months leading up to admission. Also, several months prior to admission the patient had been at a campground with known Lyme-positive ticks, but both patient and mother denied any known bites or rashes. On admission, neurological exam was normal. Electroencephalogram was within normal limits for age. Lumbar puncture demonstrated an opening pressure greater than 55-cm H2O, but an extensive CSF work-up was negative. Susceptibility weighted imaging (SWI) magnetic resonance imaging (MRI) revealed bilateral peripapillary intraretinal hemorrhages. Optic nerve head protrusion and enhancement were noted on post-contrast T1-weighted MRI with mild flattening of the posterior globes. Narrowed transverse sinuses were seen on 2-dimensional time of flight magnetic resonance venogram. Fundoscopic examination revealed bilateral papilledema with splinter peripapillary hemorrhages. Formal visual field testing revealed a paracentral scotoma consistent with optic nerve swelling (Figs. 1 and 2).
Retinal hemorrhages (RHs) are widely known to be one of the primary manifestations of non-accidental trauma (NAT) with the incidence increasing for more severely injured infants. Repeated acceleration-deceleration forces appear to play a primary role in the development of orbital damage in NAT largely because of the shearing forces at points of attachment. Increased intracranial pressure has also been suggested as an etiology for RH; however, this theory has received many arguments against it, the most prominent of which is the lack of RH in other causes of increased intracranial pressure. RHs are relatively infrequently demonstrated in idiopathic intracranial hypertension (IIH), and when present, they are intraretinal and located in a peripapillary pattern that is distinct from the multilayered and widespread pattern of RH in NAT.[3, 4] As previously demonstrated, RH may be easily depicted with SWI in the setting of NAT. SWI uses a different type of contrast in MRI that is different from spin density, T1, or T2 imaging that allows the exploitation of the susceptibility difference between tissues.
SWI is sensitive to deoxyhemoglobin and is therefore useful for visualizing venous blood in hemorrhages. This case demonstrates that the objective morphology and signal intensity changes in the peripapillary region of the retina on SWI may be used to detect hemorrhages in the setting of IIH. Thus, we believe the role of SWI in IIH should be investigated as it could be used as a new neuroimaging criterion for the diagnosis of IIH in the pediatric population.