Neoral is a new microemulsion formulation of cyclosporin A (CsA) that has been reported to have better absorption characteristics than sandimmune. We converted 25 long-term pediatric renal transplant recipients with a mean age of 14.1 yr and a mean follow-up period of 6.4 yr from sandimmune (SIM) to neoral (NEO) on a 1:1 basis. The mean dosage of SIM or NEO required to maintain 'therapeutic range' steady-state trough levels between 100 and 200 ng/mL was similar. We compared 6-h CsA pharmacokinetic profiles taken approximately 6 months after the conversion to NEO with the previous SIM profiles of the same patients. Generally, in the NEO profiles the time to reach the maximum concentration was shorter and the maximum concentration was higher, showing a rapid decline towards the trough-level when compared to the previous SIM profiles. During intake of NEO the AUC0-12 h in the 12-h profiles correlates strongly with the AUC0-6 h in the 6-h profiles (r = 0.98), a similar finding to that which we reported previously for SIM. The median AUC0-6 h for NEO demonstrates a 70% increase compared to the median AUC0-6 h for SIM. Despite the increased drug exposure NEO was well tolerated and did not cause any apparent toxicity within the first 6 months after conversion. The CsA blood level 2 h after intake of NEO showed a higher correlation with the AUC0-12 h (r = 0.91) than the trough level (r = 0.64). The abbreviated profile based on three early sampling points and calculated by AUCPRED = 335.9 + 1.1*(C1) + 1.1*(C2) + 5.4*(C4) correlated well with the full AUC (r2 = 0.98, p < 0.0001). Mean prediction error (+/- SD) was 0.16% (+/- 4.32), and in no patients did the calculated values fall outside the 10% prediction error limit. We therefore conclude that NEO exhibits a higher bioavailability in children compared to SIM without causing apparent toxicity. Monitoring of the C2 might be a better alternative for trough level monitoring in daily clinical practice. A strategy of three early sampling points (C1, C2 and C4) allows a reliable AUC0-12 h prediction and can reduce the length of observation, making it a useful and cost-effective tool in clinical practice.