To determine which hydration (saline, saline + mannitol, or saline + furosemide) is associated with least cisplatin nephrotoxicity.

We randomized 49 women who received cisplatin (75 mg/m(2) every 3 weeks) into one of the three hydration arms. The 24-h creatinine clearance was measured before and on day 6 after cisplatin infusion. The patients of each arm received 2 l of saline hydration. In the saline + furosemide arm, 40 mg of furosemide was given after hydration. In the saline + mannitol arm, 50 g of mannitol was mixed with the cisplatin.

For the first cycle of chemotherapy, 15 women were randomized to saline, 17 to saline + furosemide, and 17 to saline + mannitol. For each group, the creatinine clearances before cisplatin infusion were (means+/-SD, milliliters per minute) 84.5+/-26.8, 82.5+/-24.0 and 87.4+/-25.6, and after cisplatin infusion were 79.1+/-31.9, 68.7+/-21.5, and 56.4+/-22.9, respectively. The decreases in creatinine clearance were similar between the saline group and the saline + furosemide group ( P=0.66), but different between the saline + mannitol group and the saline group ( P=0.02) or the saline + furosemide group ( P=0.02). As each woman received multiple courses of cisplatin, 15 who received saline contributed 41 paired datasets, 17 who received saline + furosemide contributed 49 paired datasets, and 17 who received saline + mannitol contributed 36 paired datasets showed similar patterns.

Hydration with saline or saline + furosemide appears to be associated with less cisplatin nephrotoxicity than saline + mannitol.