Objective To evaluate the feasibility, complications and preliminary results of laparoscopic pelvic lymphadenectomy and radical vaginal hysterectomy for the treatment of early invasive carcinoma of the cervix. Design A retrospective randomized clinical study, Setting The Gynaecology Oncology Unit at the National Women's Hospital, Auckland, New Zealand. Subjects These were 30 women who had presented to the unit with early stage cervical cancer for whom surgery was thought to be the most appropriate primary treatment. Interventions The first 15 consecutive patients who were diagnosed in 1997 (group A) underwent a standard class 2 abdominal radical hysterectomy and pelvic lymphadenectomy The first 15 consecutive patients who were diagnosed in 1998 (group B) underwent a laparoscopic pelvic lymphadenectomy and a vaginal radical hysterectomy. Main outcome measures These included the clinical factors of operating times, blood loss, complications and hospital stay, and the pathological factors of lymph node yield, surgical margins and residual tumour size. Results The operating time was slightly longer in group B (mean = 3.18 h). There were no complications in either group but blood loss was halved in group B (mean = 353 mi), Hospital stay was also shorter (mean = 6 days) compared with group A (mean = 8 days). Lymph node yield was comparable to that obtained in open surgery (mean = 18.1 nodes) and clear margins were obtained in 14 out of 15 patients. Conclusion Our preliminary results demonstrate the feasibility and safety of laparoscopic pelvic lymphadenectomy and vaginal radical hysterectomy as an alternative to open radical hysterectomy and pelvic lymphadenectomy.