Cyclosporin A (CsA) absorption is highly variable in BMT patients. Neoral, a new microemulsion formulation of CsA, permits increased absorption with less variable pharmacokinetic parameters in non-BMT patients. We evaluated the pharmacokinetics of CsA after BMT in patients received microemulsion CsA. Two oral doses of 3mg/kg were given 48 h apart between 14 and 28 days after allogeneic BMT in 20 adults, and one dose in seven children, while subjects were receiving a continuous i.v. infusion of CsA. Whole blood samples were taken throughout the dosing interval to calculate the incremental CsA exposure using maximum concentration (Cmax), time to Cmax (tmax), concentration at 12 h after the dose (C12), the area under the concentration-time curve (AUC), and to establish inter- and intra-patient pharmacokinetic variability. Drug exposure was substantially lower in children than adults, with an AUC of 861+/-805 vs 2629+/-1487 micromg x h/l (P = 0.001), respectively, and absorption was delayed and diminished in both groups by comparison with solid organ recipients. Intra-patient variability in adults for AUC was high at 0.59+/-0.34, while inter-patient variability, measured as the coefficient of variation (c.v.), was 0.55 for the first and 0.54 for the second dose. In adults, gastrointestinal (GI) inflammation due to either mucositis or GVHD resulted in a higher AUC of 3077+/-1551 microg x h/l compared to 1795+/-973 microg x h/l (P = 0.02), and a similar trend was observed in children. AUC seemed little affected by the CsA formulation (liquid or capsule), or co-administration with liquids or food. Trough (12 h) CsA levels correlated poorly with incremental AUC. Sparse sample modeling of the AUC using two-point predictors taken at 2.5 and 5 h after dosing accurately approximated AUC in adults (r2 = 0.94), while 1.5 and 5 h was superior in children (r2 = 0.98). These data suggest that 12 h postdose trough measurements of CsA may not be the most appropriate way to evaluate CsA blood concentrations in order to establish therapeutic efficacy in BMT patients. Based on this study, the dose of microemulsion CsA should be adjusted based on recipient age, and the presence of GI inflammation secondary to mucositis or GVHD. These data would suggest that sparse sampling at time points earlier than the trough more accurately reflects the AUC and may correlate more closely with therapeutic efficacy early post-BMT.