In order to determine if routine yearly evaluations of children with shunted hydrocephalus were likely to diagnose shunt malfunction, we reviewed the medical records of the last 100 children who had such routine evaluations. Only 4 children had symptoms that were potentially referable to malfunction; none were subsequently found to have malfunction and no child had signs of malfunction. We also reviewed the medical records of the last 100 children who had shunt revisions to determine if any of them were diagnosed during routine follow-up examinations. Four were: 2 with clinical signs of malfunction and 2 with evidence of malfunction on routine scans. Only 1 child in either group had signs or symptoms of malfunction if they had a functioning shunt in place for longer than 1 year. Although yearly follow-up visits are common practice, we conclude that such examinations are unlikely to detect shunt malfunction. Follow-up at intervals of 2 years is probably appropriate after 2 years of age if the child has had a functional shunt for 1 year and has had a scan that indicates a functioning shunt.