PURPOSE: Liposomal amphotericin B (L-AMB) is an essential antifungal agent for patients with hematologic diseases; however, the drug causes severe hypokalemia at a high frequency. Meanwhile, there is little evidence regarding the risk factors for L-AMB-induced severe hypokalemia, and the prevention protocol has not been established. The goal of this study was to identify the risk factors related to severe hypokalemia induced by L-AMB in hematologic patients. METHODS: Seventy-eight hematologic patients with a first administration of L-AMB were enrolled in the study. Eleven patients who had serum potassium levels <3.0 mmol/L before L-AMB administration and 12 patients who received L-AMB administration within 3 days were excluded. Patients who had a serum potassium level <3.0 mmol/L during L-AMB administration were classified into a hypokalemia group (n = 26), and those who had a serum potassium level ≥3.0 mmol/L were classified into a non-hypokalemia group (n = 29). The patient characteristics were analyzed retrospectively. In addition, the usefulness of potassium supplementation was analyzed for those patients who received potassium formulations (non-hypokalemia group, n = 15; hypokalemia group, n = 24). FINDINGS: Twenty-six patients had hypolalemia after L-AMB administration. Hypokalemia with serum potassium levels <3.0 mmol/L was observed ~7 days after starting L-AMB administration. The patient characteristics, L-AMB dose, and L-AMB administration period did not differ between the 2 groups. In the patients who received potassium formulations, the period between starting L-AMB administration and starting potassium supplementation was significantly shorter in the non-hypokalemia group than in the hypokalemia group (median, 0 vs 4 days, respectively; P < 0.01); the potassium dose was not different between the 2 groups. A receiver-operating characteristic curve revealed that the cutoff time for the start of potassium supplementation to reduce the incidence of L-AMB-induced hypokalemia was 3 days. Multivariate logistic regression analysis revealed that beginning potassium supplementation within 2 days from the start of L-AMB administration was an independent factor reducing the risk of L-AMB-induced hypokalemia (odds ratio, 0.094 [95% CI, 0.019-0.47]). IMPLICATIONS: This study showed that starting administration of a potassium formulation within 2 days from the start of L-AMB administration was a risk reduction factor for L-AMB-induced hypokalemia. This finding indicates that early potassium supplementation should be incorporated into the regimen of hypokalemia management when L-AMB is used.