Forty-nine studies were included (n=7157 stated in text, 5924 stated in table).
Numbers given are events per patient.
Rejection: all stable transplant recipients has a similar incidence of rejection comparing Neoral and Sandimmune (0.06 versus 0.07, p = not significant). Subgroup analysis showed no significant differences in rejection incidence comparing Neoral and Sandimmune in stable renal and liver patients. Rejection was higher in de novo transplant recipients in the Sandimmune group (0.51 versus 0.37, p<0.05). In de novo renal transplant recipients rejection was also higher in the Sandimmune group (0.46 versus 0.34, p<0.05) and the same was seen in de novo liver transplant recipients (0.59 versus 0.43, p<0.05).
Adverse events: All stable transplant recipients had similar numbers of adverse events comparing Neoral and Sandimmune (1.04 versus 1.04, p = not significant). Subgroup analysis of stable renal and liver recipients showed no significant differences. De novo transplant recipients had similar numbers of adverse events comparing Neoral and Sandimmune (4.58 versus 4.83, p = not significant). De novo renal transplant recipients had similar numbers of adverse events (1.37 versus 1.38). De novo liver recipients had twice as many adverse events on Sandimmune than on Neoral (14.2 versus 7.1, p<0.00001).
Graft loss: similar in the Sandimmune and Neoral de novo and stable patients in both kidney and liver recipients and did not reach statistical significance in any of the studies.
Serum creatinine levels: the majority of the studies did not record serum creatinine levels. In most cases the authors stated that no difference in renal function was observed when comparing Neoral and Sandimmune.