Diabetes insipidus is a symptom which may be due to a variety of causes. Its diagnosis and treatment, unsatisfactory as the latter may be, present no difficulty, but the detection of the underlying cause often requires considerable study and may at times be impossible. When this underlying cause is not ascertainable a diagnosis of idiopathic diabetes insipidus may be made. In many instances this classification will be correct, since neither the further course nor the necropsy will reveal an anatomic cause for the polydipsia and polyuria. In other cases the basic process will be recognized only after the symptoms have progressed for months or years.The observation of a few cases of polyuria and polydipsia and the attempt to formulate their prognoses induced us to study the literature of diabetes insipidus in children. This survey demonstrated that definite progress has recently been made in the understanding of some of the etiologic processes underlying this condition and also that its classification as well as the evaluation of its symptomatology should be recon¬ sidered. No attempt has been made to compile all the reported cases of diabetes insipidus. Each section is summarized, so that a final sum¬ mary appears to be unnecessary. From the Department of Pediatrics, University of Cincinnati, and the Children's Hospital Research Foundation.In this paper considerable space has been given to therapy with extracts of the posterior lobe of the pituitary gland. There is much confusion in the terminology employed by various authors in regard to these extracts. This arises partly because much of the material quoted was written before a distinction had been made between pitressin and pitocin. It is possible that some of the experimental studies and clinical results are confused because proper extracts were not employed. In this review an attempt has been made to specify the type of extract administered and when possible to use the terminology employed by the respective authors. Obviously the active antidiuretic substance is pitressin. Downloaded From: http://archpedi.jamanetwork.com/ by a University of Manitoba User on 06/16/2015 PATHOGENESIS The pathogenesis of diabetes insipidus is not completely under¬ stood.1 Postmortem studies early directed attention to the region of the pituitary gland and the adjacent nerve centers in the floor of the third ventricle (Erdheim). It would appear that the hypophysis and the diencephalic region not only are closely related anatomically but con¬ stitute a functional unit (Biedl). Thus far attempts to attribute diabetes insipidus to any single part of this system have failed of general recognition.The theory that disturbance of the posterior lobe of the pituitary gland is operative in causing the symptoms of diabetes insipidus is based on evidence obtained experimentally, clinically and at necropsy.2 It is best supported by the fact that in the majority of cases the symptoms disappear or are modified after administration of extracts of the posterior lobe of the pituitary gland.3 There a...