Magnesium supplementation has been reported to have a nephroprotective effect on cisplatin-induced renal dysfunction, but little evidence exists regarding the effect of magnesium preloading before cisplatin administration. We started to include magnesium preloading (8 mEq) in cisplatin-containing treatment regimens in January 2011. The aim of the present study was to evaluate whether magnesium preloading reduces cisplatin-induced nephrotoxicity. We retrospectively reviewed 496 thoracic malignancy patients treated with cisplatin (≥60 mg/m²)-containing regimens as a first-time chemotherapy between January 2009 and December 2011. We compared the incidence of Grade ≥2 serum creatinine elevation according to the Common Terminology Criteria for Adverse Events, version 4.0, between magnesium preloading group (n = 161 [32%]) and non-magnesium preloading group (n = 335 [68%]) during the first cycle and all cycles. The median number of administered cycles was four in both groups. The incidence of Grade ≥2 serum creatinine elevation in magnesium preloading group was significantly lower during both the first cycle and all cycles than in the non-magnesium preloading group (4.9 versus 19.1% during the first cycle, and 14.2 versus 39.7% during all the cycles). A multivariate analysis indicated that magnesium preloading significantly reduced cisplatin-induced nephrotoxicity throughout the entire period from after the first administration (odds ratio: 0.262, 95% confidence interval: 0.106-0.596 during the first cycle, and odds ratio: 0.234, 95% confidence interval: 0.129-0.414 during all cycles). Magnesium preloading before cisplatin administration significantly reduced cisplatin-induced nephrotoxicity.