Aicardi subject 1 is a 10-year-old girl who was born at 38 weeks to a 31-year-old gravida 2 mother. Parents were nonconsaguinous. A 30-week ultrasound suggesting hydrocephalus resulted in a postnatal MRI, which revealed agenesis of the corpus callosum. At birth the child’s weight was 3374g. The child remained in the hospital for 7 days because of jaundice.
Bilateral chorioretinal lacunae were documented at 3 months. The left macula lacunae were more severe than the right and she had left eye esotropia. Magnetic resonance imaging revealed complete callosal agenesis and type II interhemispheric cysts [14
], which were subsequently fenestrated. She also had colpocephaly, bilateral subependymal nodular heterotopia, bilateral subcortical heterotopia, left frontal polymicrogyria, and an absent anterior commissure. In the posterior fossa, arachnoid cysts and cerebellar dysplasia were noted.
The child developed infantile spasms in the first few months of life and these occurred two times per day upon waking from sleep. She later developed tonic seizures, more severe clonic seizures, and gelastic seizures. Her antiepileptic regimen included clorazepate, lamotrigine, levetiracetam, and diazepam rectal gel as needed. She had spastic quadriplegic cerebral palsy. In addition, she was nonverbal, did not signal, and did not follow commands, but clearly responded to her name. She laughed and smiled frequently in response to specific stimuli, used eye contact, and responded to attention. Because of her severe intellectual impairment, standardized intelligence testing was not performed.
We conducted an analysis of chromosomal copy number variation using a custom NimbleGen array with 380,000 oligos spanning the X chromosome, comparing DNA from each proband to their respective mothers. This analysis did not reveal any clinically significant copy number variations in either patient.
Aicardi subject 2 is a 34-year-old woman who was born to a 29-year-old primigravida mother. Parents were nonconsanguineous and later had 2 healthy children. The patient was delivered by normal spontaneous delivery and at birth weighed 2410g. She was not diagnosed with Aicardi syndrome until the age of 6 years, when her lacunae were identified, although infantile spasms began at 6 weeks of age.
Ophthalmological findings included chorioretinal lacunae, bilateral optic nerve colobomas, and chronic open angle glaucoma. Magnetic resonance images were limited but revealed partial absence of the corpus callosum, colpocephaly, absence of the septum pellucidum, and absence of the anterior commissure. She was hypothyroid and born with hemivertebrae, requiring a spinal fusion at the age of 9.
In the first month of life, the mother noticed her child’s abnormal tone and motor development. In addition to infantile spasms by 2 months, she developed atonic seizures. These later disappeared and she developed flexor spasms and generalized tonic clonic seizures. Her antiepileptic regimen included carbamazepine, gabapentin, levetiracetam, and diazepam rectal gel as needed. She had spastic cerebral palsy but began to walk at 5 years and used her right hand only with partial thumb opposition. She was also nonverbal and did not respond to her name. She did not signal on her own but laughed in response to specific stimuli, recognized faces, responded to attention, and used eye gaze and touch to communicate needs. Because of her severe intellectual impairment, standardized intelligence testing was not performed.
For comparison purposes, subjects with callosal dysgenesis and associated cortical malformations, such as heterotopia and/or polymicrogyria, who had undergone a similar DTI protocol were retrospectively identified. To compare with Aicardi subject 1, an eight year-old male was identified with complete callosal agenesis and subcortical and periventricular heterotopia. This control subject had an intelligence quotient (IQ) of 70, as measured by the Wechsler Abbreviated Scale of Intelligence (WASI). For comparison with Aicardi subject 2, a 34 year-old male with partial callosal agenesis, periventricular heterotopia and polymicrogyria, was identified. This control subject had obsessive-compulsive disorder and an IQ of 90, measured by the WASI. These subjects will be referred to as control subjects 1 and 2, respectively.
All subjects underwent MR imaging at our institution, including high resolution T1-weighted structural imaging and DTI at 3 Tesla. For subject 1, DTI was performed with 19 diffusion-encoding directions, TR/TE = 11500/68, with 2.2-mm axial sections, in-plane resolution of 2.2 × 2.2 mm, and a diffusion-weighting strength of b = 1000 s/mm2. For Aicardi subject 2, DTI was performed with 55 diffusion-encoding directions, TR/TE = 7000/65, with 1.8-mm axial sections, in-plane resolution of 1.8 × 1.8 mm, and a diffusion-weighting strength of b = 1000 s/mm2. Due to subject movement, the T1-weighted images for subject 2 were of low quality and not included in the study.
Control subject 1 underwent DTI imaging at 3T with 25 diffusion-encoding directions, TR/TE = 11000/81, with 2.4-mm axial sections, in-plane resolution of 2.2 × 2.2 mm, and a diffusion-weighting strength of b = 1000 s/mm2. For control subject 2, a 55 diffusion-encoding direction DTI scan was obtained with identical acquisition parameters as those used for subject 2.
DTI datasets were corrected for motion and eddy current artifact using a linear image registration tool [15
]. Color fractional anisotropy (FA) maps were then generated to visualize white matter organization using DtiStudio v2.4 software (http://www.mristudio.org
) as described elsewhere [16
]. Three-dimensional whole brain fiber tractography was performed, using a deterministic streamline algorithm [17
]: tracts were initiated from all voxels with a fractional anisotropy greater than 0.2, and terminated at voxels with a fractional anisotropy less than 0.3, or a turning angle between adjacent voxels of greater than 50°. Major white matter tracts were isolated by manually drawn regions of interest (ROIs) using standard protocols [18
]. For verification purposes, an additional tracking procedure was performed with less stringent fractional anisotropy thresholds, such that tracts were initiated from all voxels with a fractional anisotropy greater than 0.15, and terminated at voxels with a fractional anisotropy less than 0.15 or a turning angle of greater than 50°.