The epileptic semiology of 19 patients (from 15 families) with mitochondrial disease due to mutations in the POLG1 gene is presented. The patients were either homozygous for the 1399G > A (p.A467T) or 2243G > C (p.W748S) mutations or compound heterozygotes for these two mutations. While the clinical features have been reviewed, detailed analysis of their epilepsy is presented for the first time. Irrespective of genotype, patients developed an epileptic syndrome with initial features of occipital lobe epilepsy. Occipital seizure phenomena included flickering coloured light, sometimes persisting for weeks, months or even years, ictal visual loss, horizontal/vertical nystagmus or oculoclonus, dysmorphopsia, micro-/macropsia and palinopsia. Most patients developed simple partial seizure phenomena with motor symptoms suggesting frontal lobe seizure initiation or spread. Simple and complex partial seizures, clonic- and/or myoclonic seizures with epilepsia partialis continua and frequent convulsive status epilepticus were observed in this syndrome that appears to be a symptomatic and secondary generalized or multifocal epilepsy with focal occipital predilection. The mean age of seizure presentation was 18.4 years (6-58 years). All patients developed status epilepticus and 11 patient deaths were, all related to prolonged convulsive status epilepticus, including two with liver failure apparently precipitated by treatment with sodium valproate.

Pontocerebellar hypoplasia (PCH) is rarely associated with anterior horn cell disease and designated as PCH-1. This phenotype is characterized by severe muscle weakness and hypotonia starting prenatally or at birth with a life span not exceeding a few months in most cases. Milder disease courses with later onset and longer survival are normally not diagnosed as PCH-1. We describe the clinical and neuroradiological findings in nine patients out of six siblingships with evidence of cerebellar defects and early onset spinal muscular atrophy (SMA), representing a broad spectrum of clinical variability. In all patients, the diagnosis of SMA (Werdnig-Hoffmann disease) was made on the basis of electrophysiological data and muscle biopsy; however, genetic testing failed to confirm the diagnosis of infantile SMA with a gene defect on chromosome 5q and resulted in clinical reevaluation. Age at onset was after a normal period in the first months of life in three siblingships and pre- and postnatally in the other three families. Life span was 2-4 years in patients with later onset, and age at death occurred after birth or within months in the more severe group. Two siblingships showed discordant ages at death despite similar treatment. In contrast to the previous definition of PCH-1, our observations suggest the existence of milder phenotypes with pontocerebellar hypoplasia or olivopontocerebellar atrophy in combination with anterior horn cell loss. A pontine involvement is not necessarily seen by neuroimaging methods. The genetic basis of PCH-1 remains to be determined. The gene locus for infantile SMA on chromosome 5q could be excluded by linkage studies. Parental consanguinity and affected siblings make autosomal recessive inheritance most likely.

Growth and neurodevelopment at 6, 10 and 18 months of age have been studied prospectively and longitudinally in a series of 17 children born to mothers who used benzodiazepines (BZD) in therapeutic doses as their only psychotropic drug throughout pregnancy. The results were compared with a group of 29 children born to mothers without any known use of psychotropic drugs. The BZD-exposed children caught up their low mean birth-weight, at an early stage, whereas the slightly decreased head circumference at birth remained at the same low level. In five infants, a pattern of craniofacial anomalies was found. Deviating neurodevelopmental and clinical symptoms and signs were common. The gross motor development was retarded at 6 and 10 months, but was nearly normal at 18 months. Impaired fine motor functions were found on all follow-up occasions. At 18 months, the most prominent finding was a delayed development of pincer grasp. The BZD-exposed children showed deviations in muscle tone and pattern of movements more frequently than children in the reference group. The study suggests that the use of BZD in therapeutic doses throughout pregnancy can have negative effects on the development of children up to 18 months of age. The long-term hazards cannot be evaluated from these results. A further follow-up at early school age is needed and is in progress.

Seventeen infants born to mothers who used benzodiazepines (BZD) throughout pregnancy were followed prospectively and compared with 29 infants born to mothers who had not used psychoactive substances. On the Griffiths' Developmental Scale, the BZD-exposed infants demonstrated consistently lower mean GQs and DQs for all subscales at 5, 10 and 18 months of age. The differences in GQ reached statistical significance at 10 and 18 months. The DQs differed significantly for all subscales at 10 months and for the personal-social behaviour and hearing and speech subscales at 18 months of age. We suggest that prenatal exposure to BZD may cause a general delay in mental development up to 18 months of age.