OBJECTIVE: The aim of the study was to compare the initial surgical outcomes and learning curve of nerve-sparing robotic radical hysterectomy (RRH) with nerve-sparing total laparoscopic radical hysterectomy (TLRH) for the treatment of early-stage cervical cancer in the first 50 cases. METHODS: Between January 2008 and March 2012, 50 consecutive patients underwent nerve-sparing RRH. These patients were compared with a historic cohort of the first 50 consecutive patients who underwent nerve-sparing TLRH. RESULTS: Both groups were similar with respect to patients and tumor characteristics. The mean operating time in the RRH group was significantly longer than that in the TLRH group (230.1 ± 35.8 vs 211.2 ± 46.7 minutes; P = 0.025). The mean blood loss for the robotic group was significantly lower compared with the laparoscopic group (54.9 ± 31.5 vs 201.9 ± 148.4 mL; P < 0.001). There was no significant difference in the mean pelvic lymph nodes between the 2 groups (25.0 ± 9.9 vs 23.1 ± 10.4; P = 0.361). The mean days to normal residual urine were 9.6 ± 6.4 in RRH and 11.0 ± 6.2 in TLRH (P = 0.291). The incidence of intraoperative complication was profoundly lower in RRH compared with that of TLRH (0% vs 8%; P = 0.041). Moreover, no intraoperative transfusion was required in RRH, whereas 4 (8%) were required in TLRH (P = 0.041). In both groups, we found no evidence of a learning effect during the first 50 cases. CONCLUSIONS: During the first 50 cases, surgical outcomes and complication rates of nerve-sparing RRH were found to be comparable to those of nerve-sparing TLRH. Moreover, the mean blood loss and intraoperative complication rate in the robotic group were significantly lower than those in the laparoscopic group. Surgical skills for nerve-sparing TLRH easily and safely translated to nerve-sparing RRH in case of experienced laparoscopic surgeon.