A comparison of diagnostic methods to differentiate diabetes insipidus from primary polyuria: a review of 21 patients.

Direct measurement of plasma AVP and indirect assessment of antidiuretic activity during standard dehydration tests were made in 21 polyuric and polydipsic patients to establish the efficacy of each method in determining the cause of polyuria. Patients with acquired nephrogenic diabetes insipidus (e.g. diabetes mellitus, renal failure, hypercalcaemia) were excluded from the study. Cranial diabetes insipidus was diagnosed by plasma AVP responses to osmotic stimulation during infusion of hypertonic 5% saline which were subnormal in 13 patients, 4 of whom had undetectable plasma AVP and 3 who had reduced but osmoregulated AVP release. Standard water deprivation tests confirmed cranial diabetes insipidus in all but 2 patients who were diagnosed as partial nephrogenic diabetes insipidus. The remaining 8 patients had normal, osmoregulated AVP secretion; the cause of their polyuria was determined by their renal response to desmopressin. Two patients had nephrogenic diabetes insipidus and 6 had primary polydipsia. The majority of polyuric patients could be accurately diagnosed by carefully performed dehydration tests. We suggest that direct measurements of plasma AVP during osmotic stimulation are only necessary to distinguish mild forms of cranial from nephrogenic diabetes, or to define precisely the characteristics of AVP secretion.