The aim of this study was to perform a quantitative meta-analysis of the average bioequivalence criteria between Sandimmun and Sandimmun Neoral in kidney transplant patients, and to review the new bioequivalence criteria and their application to generic formulation of cyclosporin. In Medline, we searched for clinical trials evaluating the bioequivalence between Sandimmun and Sandimmun Neoral in kidney transplant patients and we collected the information regarding the bioequivalence, study design, sample size, and time post-transplant. The effect was measured by the Wolf method; publication bias was evaluated by the Galbraith method and the Rosenthal formula was used to calculate the number of additional studies with no statistical differences needed to get a statistically non-significant overall estimation. We selected 6 clinical trials with a latin square design and 4 clinical trials with sequential design. The average bioequivalence criteria between Sandimmun Neoral and Sandimmun were 1.327 (90% CI: 1,311 a 1,344), 1,663 (90% CI: 1,635 a 1,692) and 0.559 hours (90% CI: 0.544 a 0.574 hours) for logharitmic transformation of area under the curve and maximum concentration, and time to maximum concentration, respectively. For these three outcomes, we found statistical differences between different study designs and for area under the curve and maximum concentration, the average bioequivalence criteria significantly fall with the post-transplant time. We conclude Sandimmun Neoral and Sandimmun are not bioequivalents and the experience reached with these two drugs is not applicable to the evaluation of generic formulations of cyclosporin.