PILOT STUDY: AN ETIOLOGIC CLASSIFICATION OF PERVASIVE DEVELOPMENTAL DISORDERS

Background: Pervasive Developmental Disorders (PDD) is etiologically heterogeneous. Objectives: To classify children diagnosed with PDD according to organic diagnoses. Methods: Comparable with epilepsy classification, we separated 436 children diagnosed with PDD into three groups: Symptomatic - a diagnosed organic- neurologic disorder was identified Cryptogenic - an underlying etiology was suspected Idiopathic- without evidence of other neurological disorders Children were recruited from an Israeli (N=351) and American (N=85) databases. Results: American database: Nine children were classified as symptomatic (44%males), 23 cryptogenic (74%males) and 53 idiopathic (81%males). PDD subtype also differed between the groups - 53% idiopathic, 43% cryptogenic and 11% symptomatic meeting full criteria for Autism. Among other findings, there was evidence for higher rates of language delay (89% vs. 66%) and no evidence for regression in the symptomatic group, but similar rates of each between idiopathic and cryptogenic groups (21-34%). Israeli database: Twenty two children were classified as symptomatic (36.3%males), 92 cryptogenic (86.9%males) and 237 idiopathic (79.3%males). After excluding the Rett syndrome (N=8) from the symptomatic group, the percentage of males and females became almost equal (57.1% males). PDD subtype also differed between the groups - 43% idiopathic (44.1% vs 38.7% in males and females respectively), 47.8% cryptogenic (47.5% vs 50% in males and females respectively) and 42.8% symptomatic (37.5% vs 50% in males and females respectively) meeting full criteria for Autism. Among other findings, the entire symptomatic group showed language delay but almost none had evidence for regression. Similar rates of each were seen in both idiopathic and cryptogenic groups (above 90% had language delay, and a fifth showed evidence for regression). Conclusions: In this study, differences in gender, clinical and diagnostic features were found when etiology was used to create subtypes of PDD. This classification could have heuristic importance in search for an autism gene(s).