A Clinical Study of Secretin in Autism and Pervasive Developmental Delay

Purpose: The aim of this study was to examine the clinical effect of a single dose of secretin given to patients whose disease status lay within the autistic spectrum. Design: Open, uncontrolled trial. Materials and Methods: Sixty-four patients were registered for the study. They included 3 female adults and 2 male adolescents. There was a total of 16 dropouts. Thus there were 48 evaluable patients. Each patient received a complete vial of secretin, irrespective of current age and body weight. Any existing treatment that had already been started before the study began was allowed to continue. No new treatment was permitted for a minimum period of 8 weeks when study surveillance was completed. The parents or guardians performed clinical assessment on a weekly basis, using numerical scoring of symptoms in a questionnaire. The only laboratory study performed was a hair analysis at the outset and at the completion of the 8 weeks of surveillance. Results: Of 48 evaluable patients, there were 39 whose clinical response varied from minor to major behavioral and/or bowel function improvement and 9 in whom there was no sign at all of either a behavioral or gastrointestinal response. Only a few of these patients showed significant abnormalities in hair analysis and there was no significant difference in these changes as compared with a group of children who had been seen at our clinic for reasons other than autistic spectrum. The side effects observed were temporary irritability in a few patients and exacerbation of previously quiescent seizures in one child. Conclusions: We found a clearcut population of clinical ''responders'' as compared with a much smaller group of ''non-responders.'' Little can be said about the dropout patients other than to conclude that total lack of response and expense were the most likely causes. We can certainly encourage and condone the use of secretin as a valuable weapon in attempting treatment of this devastating spectrum of disease. This study has not helped to define the indications to suggest a ''responder'' from a ''non-responder.''