Thirty patients with intractable partial epilepsy and a pathological or radiological diagnosis of neuronal migration disorders (NMD) were studied at the Montreal Neurological Institute and Hospital, Montreal, Quebec, Canada. Age at onset of epilepsy ranged from 4 months to 21 years (mean 5.7). Prenatal etiological factors in 8 patients included maternal x-ray exposure or trauma, diabetes, advanced maternal age, uterine abnormality. All patients had partial epilepsy: complex partial (67%), partial motor (73%), and secondary generalized seizures (73%). 8 patients (27%) had tonic or atonic drop attacks, and 6 of these had lesions involving the central region. Status epilepticus had occurred in 30%. A below average IQ was found in 1/2 of the patients. MRI was superior to CT, but did not distinguish between pachygyria, cortical dysplasia or tuberous sclerosis. White matter subcortical abnormalities were demonstrated more frequently than cortical abnormalities. In 22 of 26 surgically treated patients, the histological diagnosis was focal cortical dysplasia (12), and forme fruste of tuberous sclerosis (10). Those with a histological picture of tuberous sclerosis had none of the classical clinical diagnostic signs, but 80% had delayed psychomotor development and all 10 patients had multilobar unilateral or bilateral EEG spiking. These findings differed from patients with focal cortical dysplasia whose developmental milestones were normal and epileptogenic discharges were confined to one lobe or area of the brain. 
COMMENT. The results of surgical treatment of the above 26 patients are reported in the same journal . Good or excellent seizure control was achieved in 42% and moderate control in 25%. The extent of lesion removed was most strongly correlated with the surgical outcome; 77% receiving complete or 50% excision of the lesion had a good prognosis. When excision of epileptogenic tissue was based on scalp EEG and electrocorticography studies there was no correlation with surgical outcome. The lack of specificity of the MRI in the diagnosis of tuberous sclerosis in this study is also alluded to by Fryer AE, Geneticist at the Royal Liverpool Hospital, England in an editorial concerning tuberous sclerosis . Although MRI is more sensitive than CT, CT is more specific and potentially less confusing in interpretation. A thorough clinical examination including ophthalmoscopy and ultraviolet light examination of the skin is essential to exclude the diagnosis in anyone suspect or at risk. Prenatal diagnosis is only possible by fetal echocardiography but this examination is unreliable before 26 weeks and has a high rate of false negatives.