In the randomized trial AML-BFM 93 we compared 60 mg/m2/day daunorubicin with 12 mg/m2/day idarubicin for 3 days each, combined with cytarabine and etoposide during induction. Results showed a significant better blast cell reduction in the bone marrow on day 15 in patients of the idarubicin arm (25 of 144 = 17% of patients with > or = 5% blasts compared to 46 of 149 = 31% of patients after daunorubicin, Pchi2 = 0.01). This was, however, mainly seen in high risk patients treated with idarubicin (19% vs 38%, Pchi2 = 0.007). Cardiotoxicity, WHO grade 1-3 shortening fraction reduction after induction occurred in 6% patients in both arms. Bone marrow toxicity differed slightly with a median recovery time of neutrophils >500/microl of 25 days (daunorubicin) compared to 27 days (idarubicin), P = 0.05. In the total group of patients probabilities of 5 years event-free survival and disease-free survival were similar for patients treated with daunorubicin or idarubicin (49% +/- 4% vs 55% +/- 4% and 57% +/- 4% vs 64% +/- 4%, P logrank 0.29 and 0.15, respectively). However, in patients presenting with more than 5% blasts on day 15 there was a trend for a better outcome after treatment with idarubicin (P logrank 0.06). Together with the early effect seen for high risk patients these results indicate a better efficacy of idarubicin than of daunorubicin during induction with a similar rate of toxicity.