Benign familial and nonfamilial forms of infantile seizure and partial epilepsies of infancy are significant categories of pediatric epilepsies. In the past, seizures with onset during the first months of life were considered to have a bad prognosis and a symptomatic etiology. Since the first description by Fukuyama, the existence of infantile seizures with a benign evolution has been defined and accepted.

The international classification of epilepsies and epileptic syndromes¹⁷ includes benign infantile seizures, which are divided into familial and nonfamilial forms. In recent years, variants of these two forms and similar entities have been described.

In 1963, Fukuyama was the first to report cases with onset within the first 2 years of life that were characterized by generalized convulsions, absence of etiologic factors, and benign outcome.¹⁸ This type of infantile convulsion was the subject of many subsequent studies, but only after 20 years were the true clinical entities clarified. Reports have specified the localization and semiology of seizures, which have been defined as being of partial type^69,71,72 and in terms of the presence or absence of familial occurrence.^6,7,8

Vigevano and coworkers focused on cases exhibiting a family history of convulsions with benign outcome during infancy and autosomal-dominant inheritance, and they proposed the term benign infantile familial convulsions (BIFC).⁶⁷ Later, other such cases were reported in many different parts of the world,^{16,21,35,40,45} thus confirming them as new epileptic syndromes. When these entities were included in the list of epileptic syndromes by the International League Against Epilepsy (ILAE) Task Force on Classification and Terminology, it was suggested that the term “seizure” should be used rather than the term “convulsion.”¹⁷ Just like benign seizures with neonatal onset, benign infantile seizures are divided into familial and nonfamilial forms.¹⁷ These two forms, however, can have overlapping features.³⁹ Genetic studies of familial forms led to the identification of a chromosome marker on chromosomes 19,²³ 16,¹¹ and 2.⁴²

An association between benign familial infantile seizures (BFIS) and variably expressed paroxysmal choreoathetosis was reported in 1997 by Szepetowski et al.⁶¹ A specific marker on chromosome 16 has been identified in this familial variant, called infantile convulsions and choreoathetosis (ICCA).

Heron et al.²⁵ in 2002 and later Berkovic et al.³ proposed a new entity and coined the term benign familial neonatal-infantile seizures (BFNIS) to describe families with onset at an intermediate age between neonatal and infantile forms.

Considering that convulsive manifestations are limited to a short period of time, some authors hypothesized the existence of particular etiologic factors in some sporadic cases such as cases of benign infantile convulsions associated with episodes of diarrhea caused by rotavirus infections.^12,28,30 Finally, Capovilla et al. described a peculiar form of benign epilepsy occurring within the second year of life and with a typical sleep electroencephalographic (EEG) pattern.^5,6

Benign partial seizures in infancy are a group of diseases characterized by onset during the first 2 years in otherwise normal children. They could be familial, with a characteristic autosomal-dominant trait of inheritance and a typical onset around 6 months, or nonfamilial, which usually occur later. Seizures are partial with or without secondarily generalization and are typically grouped in clusters of many per day. In most cases, interictal EEGs are normal, except for the midline spikes during slow sleep described by Capovilla, but this form of epilepsy is under discussion. Outcome is always excellent, with a normal psychomotor development after seizures.

Several cases with this syndrome have been described from all over the world. Data on prevalence and incidence are not available. In a series described by Caraballo et al.,⁹ benign infantile seizures were listed as the third-most-common type of epilepsy in the first 2 years of life.

Autosomal-dominant transmission is evident in BFIS. Due to the close similarity to benign familial neonatal seizures (BFNS), researchers first tried to find the chromosome markers described in this latter syndrome.^37,55,57 In 1994, Malafosse et al.⁴³ demonstrated that BFIS is not an allelic form of BFNS, excluding the marker on chromosome 20.

In 1997, linkage analysis was carried out in five Italian BFIS families, and a locus was mapped on chromosome 19q12–13.1 between markers D19S49 and D19245.²³ Gennaro et al.²⁰ later conducted a linkage analysis of seven families of Italian origin and demonstrated the presence of linkage to chromosome 19q in a single family, suggesting genetic heterogeneity within the examined families. Studies of familial cases with ICCA are particularly interesting. Szepetowski et al.⁶¹ demonstrated linkage to the pericentromeric region of chromosome 16 in the families with this syndrome. This finding was then confirmed by Lee et al.³⁶ in a family of Chinese origin. In 2001, Caraballo et al.¹¹ found linkage on chromosome 16p12-q12, the same region as
for ICCA, in seven families with only benign familial infantile seizures. There was a previous report describing a large family affected by paroxysmal kinesigenic dyskinesia without infantile convulsions with linkage to the ICCA region.³³ Therefore, Caraballo et al. hypothesized that chromosome 16p12-q12 is a major genetic locus underlying both benign familial infantile seizures and paroxysmal dyskinesias.¹¹ Weber reported similar results in 14 families with benign familial infantile seizures without paroxysmal choreoathetosis.⁷³

In 2001, Malacarne et al.⁴² mapped a novel locus to chromosome 2q24 in eight Italian families, thus demonstrating a genetic heterogeneity, as in other autosomal-dominant idiopathic epilepsies.

In cases described as having benign familial neonatal-infantile seizures—an intermediate form between BFIS and benign familial neonatal seizures—a missense mutation in the SCN2A gene has been found, leading to the hypothesis of the existence of a third form,²⁰ although recently a similar mutation of the same gene has been described in a family with clinical features typical of BFIS.⁵⁸

A particular etiologic factor has been recognized in some infantile seizures associated with mild gastroenteritis, with positivity to the rotavirus antigen. In the other nonfamilial forms, no clear etiologic factors have been described.