New immunosuppressants are said to be superior to cyclosporine due to their higher incidence of steroid sparing and to the reduced incidence of side-effects. From May 1992 to February 1995, 79 adults underwent primary liver transplantation using cyclosporine A (Sandimmun)-based triple drug immunosuppression. Nine patients who died early after liver transplantation due to reasons unrelated to immunological problems were excluded from this analysis. The long-term outcome of the remaining 70 patients was prospectively studied in relation to steroid and azathioprine withdrawal. They were re-evaluated 6-monthly in relation to liver and kidney function; cholesterolemia, infection, de novo diabetes mellitus and arterial hypertension, malignancy, ophthalmological and osteomuscular diseases. In case of rejection occurring during or after steroid tapering, patients were switched, by protocol, to tacrolimus therapy. Median follow-up was 81 months (range 60-96). Forty-four patients (62.8 %) were biopsied 5 years after transplant; 20 patients (28.6 %) were biopsied at a median follow-up of 32 months (range 7.8-47). Six patients (8.6 %) who refused biopsies more than 1 year after liver transplantation had normal liver values throughout the whole follow-up period. Five-year actual patient and graft survivals were 75 % and 65.8 %, respectively, for the whole group (n = 79) and 85.7 % and 74.3 % for the studied group (n = 70). Steroids could be withdrawn in all but two patients (97.1 %) at a median time of 7 months (range 3-42). Steroids were restarted in six patients (8.6 %) for extrahepatic reasons. Freedom from steroids was thus observed in 62 patients (88.6 %). Seven patients (10 %) had rejection after steroid tapering; six were switched to tacrolimus. Two patients (2.9 %) needed retransplantation because of acute and chronic rejection whilst still being on full immunosuppression. In total, three patients (4.3 %) had histological signs of chronic rejection during follow-up. At 5 years post-transplant, 66.6 % and 13.3 % of the 60 patients at risk were on cyclosporine and tacrolimus monotherapy, respectively; 93.3 % were steroid-free. Mean creatinine and cholesterol levels were 1.56 +/- 1.3 mg/dl and 193.5 +/- 56.6 mg/dl; incidences of de novo arterial hypertension, insulin dependent diabetes mellitus were 26.6 % and 13.3 %. Two patients (2.8 %) developed post-transplant lymphoproliferative disease, two (2.8 %) had skin cancer. Cyclosporine-based immunosuppression allows safe steroid withdrawal in most patients and cyclosporine monotherapy can be achieved in two-thirds without compromising graft and patient survival. Results of new immunosuppressive strategies should be approached with caution, especially when considering steroid sparing and the incidence of side-effects.