OBJECTIVE: About 30% of autistic cases experience developmental regression around 2 years of age. The clinical course and manifestations of autistic children with regression remain unclear. This study investigated the clinical features of a group of autistic children with regression. METHODS: One hundred and fifty-two children at ages of 2.5-6.5 years confirmed with autism based on DSM-IV diagnostic criteria were enrolled. They were grouped according to language development: normal or regression. The perinatal history, developmental history and characteristics of regression were investigated. The symptoms were compared between the two groups. RESULTS: Regressions were observed in 33 children (21.7%) at age of between 16 and 21 months, with loss both in communicative skills and social engagement. The regressive group was scored significantly higher on the Childhood Autism Rating Scale (CARS) (P<0.05) and had a relatively higher proportion of severely ill children (66.7% vs 45.4%; P<0.05)compared with the non-regressive group. CONCLUSIONS: Regression as a characteristic symptom occurs in some autistic children and is of value for diagnosis of autism. The autistic children with regression display more severe social and language impairments than those without regression. Regressive autism may be a special subtype.
OBJECTIVE: About 30% of autistic cases experience developmental regression around 2 years of age. The clinical course and manifestations of autisticchildren with regression remain unclear. This study investigated the clinical features of a group of autisticchildren with regression. METHODS: One hundred and fifty-two children at ages of 2.5-6.5 years confirmed with autism based on DSM-IV diagnostic criteria were enrolled. They were grouped according to language development: normal or regression. The perinatal history, developmental history and characteristics of regression were investigated. The symptoms were compared between the two groups. RESULTS: Regressions were observed in 33 children (21.7%) at age of between 16 and 21 months, with loss both in communicative skills and social engagement. The regressive group was scored significantly higher on the Childhood Autism Rating Scale (CARS) (P<0.05) and had a relatively higher proportion of severely ill children (66.7% vs 45.4%; P<0.05)compared with the non-regressive group. CONCLUSIONS: Regression as a characteristic symptom occurs in some autisticchildren and is of value for diagnosis of autism. The autisticchildren with regression display more severe social and language impairments than those without regression. Regressive autism may be a special subtype.