The results of 0KT3 use for steroid-resistant rejection rescue in adult liver transplantation were analyzed retrospectively from a single transplant center. Comparison was made with concurrent patients who had no rejection (NR) or steroid-responsive rejections (SR). The records of 290 patients who underwent 323 liver transplants from April 1985 to December 1989 were examined. The first technically successful grafts were used for this analysis (265 grafts). Follow-up was a minimum of 1 year, or until death or loss of graft. All patients received triple-drug induction immunosuppression (CsA, Aza, steroids). Initial rejection was treated with 1 g methylprednisolone bolus i.v., followed by a 5-day taper of steroids from 200 mg to 20 mg. No rejection occurred in 108 (40.8%) and SR in 86 (32.4%), and OKT3 was given for persistent rejection in 71 (26.8%).The age, sex distribution, mean follow-up, and preoperative status were similar in all three groups. The preoperative diagnoses were similar, except for fulminant liver failure, in which 19 of 20 patients experienced rejection (P<0.0001). The median hospitalization stay was 37 days for 0KT3, 27 days for SR, and 21 days for NR (P<0.0001). The median ICU stay was similar in the three groups (OKT3, 4; SR, 4; NR, 3). Infections in the first 6 weeks, and in the period of 6 weeks to 1 year posttransplant, were of similar frequency for all three groups. By the Kaplan-Meier estimation, the graft and patient actuarial survival rates were comparable. At 1 year, the graft survival rate was 79.6% for NR, 79.8% for SR, and 67.6% for 0KT3. The 1-year patient survival rate was 85.2% for NR, 83.7% for SR, and 84.5% for 0KT3. Following treatment by 0KT3, rejection was permanently reversed in 42 patients. A temporary response occurred in 12 patients, 16 patients failed to respond to OKT3, 2 patients died during therapy, and 6 of the nonresponders died within 12 months. Additional OKT3 treatment was attempted in 6 patients for persistent rejection within a 1-month interval from the previous OKT3 course. Of these 6, 4 developed lym-phoproliferative disorder, and only 1 survived in response to drastic reduction of immunosuppression. In conclusion, OKT3 was effective as rescue therapy for adult liver transplant steroid-resistant rejection. Because of the associated morbidity and expense, OKT3 should be used in a selective fashion. Failure to respond to OKT3 is a serious complication, and should not be managed by prolonged or repeated courses, but rather by alternative means.