Busulfan pharmacokinetics, specifically area under the concentration curve (AUC), have been correlated with the occurrence of veno-occlusive disease (VOD) following BMT. To evaluate the risk of VOD, we studied 66 patients who received pharmacotherapeutically monitored busulfan regimens in combination with CY, etoposide (VP16) and/or Ara-C in preparation for BMT. These patients received a total of 16 doses of busulfan dosed as 1 mg/kg/dose q 6 h beginning at 09.00 (n = 39), 18.00 (n = 2), 21.00 (n = 1) or 24.00 (n = 24) h. With the first dose, blood samples were obtained at baseline, every 15-30 min for 2 h, then every 1-2 h for 4 h. Blood was analyzed for busulfan concentration by high performance liquid chromatography and AUC calculated by the trapezoidal rule. Seventeen patients (25.8%) were not evaluable for AUC calculation due to slow absorption and/or elimination: 13 of 27 (48.1%) received the first dose between 18.00-24.00 vs four of 39 (10.2%) patients who received the first dose at 09.00 (P < 0.001). Eighteen of 51 (35.3%) evaluable patients had an AUC > 1500 mumol x min/l; 10 of whom received doses reduced proportionally to achieve an AUC = 1200 mumol x min/l starting with the 10th to 15th dose. Six of 18 (33.3%) patients with an initial AUC > 1500 mumol x min/l developed VOD vs one of 33 (3.0%) patients with an initial AUC < 1500 mumol x min/l (relative risk = 11.1; P = 0.0056). Other pharmacokinetic parameters, age, gender, type of BMT, previous therapy or pre-transplant liver function tests were not predictive of VOD. A higher incidence of VOD occurred in patients receiving BUCY (4 of 10) compared to those receiving BUCYAra-C (1 of 18) or BUCYVP16 (7 of 38), which could not be attributed to increased busulfan exposure in the BUCY patients. Routine pharmacotherapeutic monitoring of busulfan is recommended with further study to evaluate the impact of earlier and greater overall dose reduction in patients with high initial busulfan exposures.