Lithium (Li) carbonate has been reported to be able to cause some reversible functional changes in the kidney. In this study, the authors aimed to investigate whether the duration of Li treatment is the primary determinant of the changes in renal functioning due to the Li treatment. For this purpose, 10 Li-naive (mean age+/-S.D.: 34.50+/-4.85), 10 short-term (mean age+/-S.D.: 31.77+/-7.61) and 10 long-term (mean age+/-S.D.: 36.60+/-10.15) Li-treated bipolar patients were included in the study. Serum blood urea nitrogen (BUN) and creatinine, urine creatinine levels, creatinine clearance, urine osmolality before and after 8-h water deprivation and urine osmolality after desmopressin injection were measured in all patients. Serum BUN and creatinine levels were within the normal limits and not statistically different among the groups. Creatinine clearance of the long-term Li-treated group was significantly lower than both that of the Li-naive group and that of the short-term Li-treated group. After 8-h water deprivation and also after desmopressin injection, no difference was found among the groups in terms of urine osmolality. However, when each patient was evaluated individually in terms of their renal concentrating ability, partial nephrogenic diabetes insipidus was diagnosed in four patients on long-term and in two patients on short-term Li treatment. To our surprise. hypothalamic diabetes insipidus was also diagnosed in other two patients on long-term Li treatment. These results demonstrate that long-term Li treatment may cause impairment in renal concentrating ability, some of which may originate from the effects of Li on vasopressin on hypothalamic level. and a decrease in glomerular filtration rate (GFR). In the light of these data, we can conclude that long-term administration of Li may be a risk factor for Li-induced renal impairment, which is a progressive effect in nature. (C) 2001 Elsevier Science Inc. All rights reserved.