PURPOSE: To analyze survival rate and late rectal and bladder complication for patients with stage with stage I and II carcinoma of uterine cervix treated by radiation alone or combined with chemotherapy. MATERIALS AND METHODS: Between November 1984 and December 1993, 127 patients with stage I and II carcinoma of uterine cervix treated by radiation alone or combined therapy of radiation and chemotherapy. Retrospective analysis for survival rate was carried out on eligible 107 patients and review for complication was possible in 91 patients. The median follow-up was 47 months (range 3-118) and the median age of patients was 56 years (range 31-76). 26 patients were stage IB by FIGO classification. 40 were stage IIA and 41 were stage IIB. 86 cases were treated by radiation alone and 21 were treated by radiation and chemotherapy. 101 patients were treated with intracavitary radiation therapy (ICRT), of these, 80 were received low dose rate (LDR) ICRT and 21 were received high dose rate (HDR) ICRT. Of the patients who received LDR ICRT, 63 were treated by 1 intracavitary insertion and 17 were underwent 2 insertions. And we evaluated the external radiation dose and midline shield. RESULTS: Acturial survival rate at 5 years was 92% for stage IB, 75% for stage IIA, 53% for stage IIB and 69% in all patients. Grade 1 rectal complications were developed in 20 cases(22%), grade 2 were in 22 cases (24%), 22 cases (24%) of grade 1 urineary complications and 17 cases (19%) of grade 2 urinary complications were observed But no patients had severe complications that needed surgical management or admission care. Maximum bladder dose for the group of patients with urinary complications was higher than that for the patients without urinary complications(7608cGy v 6960cGy, p<0.01). Maximum rectal dose for the group of patients with rectal complications was higher than that for the patients without urinary complications (7041cGy v 6269cGy, p<0.01). While there was no significant difference for survival rate or bladder complication incidence as a function of dose to whole pelvis, Grade 2 rectal complication incidence was significantly lower for the patients receiving less than 4500cGy(6.3% v 25.5%, p<0.05). There was no significant difference between HDR ICRT group and LDR ICRT group for survival rate according to stage, on the other hand complication incidence was higher in the HDR group than LDR group. This was maybe due to different prescription doses between HDR group and LDR group. Midline shield neither improved survival rate nor deceased complication rate. The number of insertion in LDR ICRT group did not affect on survival and complication rate. CONCLUSION: In stage I and II carcinoma of uterine cervix, there was no significant difference for 5 year survival rate by radiation therapy technique. Rectal complication incidence was as a function of dose to whole pelvis and there were positive correlations of maximum dose of rectum and bladder and each complication incidence. So we recommand whole pelvis dose less than 4500cGy and maximum dose of rectum and bladder as low as possible.