Although the acute renal toxicity of cisplatin we followed renal function prospectively for a has been well documented, long-term follow-mean time of 2.3 years after arrest of cisplatin up studies in cisplatin-treated children are therapy. Chronic hypomagnesemia and moderscanty. We have evaluated the incidence and ate elevation of plasma creatinine were obcharacteristics of both acute and chronic neph-served in 6 children, hypocalciuria in 5 children, rotoxicity in 22 children (median age 8 years) and hypokalemia in 1 child. Presence of hypotreated with cisplatin as part of different chemo-magnesemia was unrelated to the total dose retherapeutic protocols. All patients exhibited a ceived or the time elapsed since cisplatin thersignificant and progressive decrease in plasma apy. Renal function studies, performed in the 6 magnesium (Mg) values soon after cisplatin ad-children with chronic hypomagnesemia, reministration. Magnesiuria also increased imme-vealed different degrees of impairment in Mg diately after therapy. Hypomagnesemia (plasma reabsorption. The functional characteristics of Mg Ο½1.4 mg/dl) occurred in 10 patients and it chronic cisplatin nephrotoxicity found in the was dose-dependent. Minimal and mean cumu-present series-contrary to prior reports-are lated doses inducing hypomagnesemia were not comparable to those present in the inherited 300 and 500 mg/m 2 , respectively. In 18 children Gitelman's syndrome.