This paper presents data obtained by questionnaires sent to local hospitals and metabolic centres in Germany, Austria, Italy and Switzerland concerning the survival and outcome of patients with urea cycle disorders treated between 1975 and 1986. A total of 130 questionnaires were sent out of which 88 questionnaires of patients were returned. This study compares results of conservative long-term management using protein restriction versus protein restriction combined with more extensive treatment (arginine/citrulline, essential amino acid supplements and sodium benzoate as alternate pathway therapy). While survival was improved in neonates receiving extensive therapy, there was an increased risk of mental retardation on long-term follow-up in this group. The time at which the first symptoms of hyperammonaemia appeared, the age at diagnosis, and the time of delay in diagnosis after the first symptom were not helpful in predicting outcome; however, when plasma ammonia concentrations exceeded 300 micromol/l initially or 480 micromol/l at its peak, none of the patients had a normal cognitive outcome. CONCLUSION: there is a need to establish improved treatment approaches and a network of "rare disease" centres to assure the rapid diagnosis and effective treatment and follow-up of affected children. This should precede any consideration of neonatal screening for urea cycle disorders.