Radiation injury to intestine following hysterectomy and adjuvant radiotherapy for cervical cancer

OBJECTIVE: To evaluate the risk factors for nonrectal radiation-induced intestinal injury (NRRIII) following adjuvant radiotherapy (RT) for cervical cancer using a retrospective review of medical records. METHODS: From September 1992 to December 1998, 164 patients with uterine cervical cancer that had completed their allocated adjuvant radiotherapy at the Chinese Medical University Hospital were enrolled for NRRIII analysis. The patients were classified into two groups according to the extent of surgery. Group A consisted of 110 patients (International Federation of Gynecology and Obstetrics [FIGO] stage: IB, n = 87; IIA, n = 21; IIB, n = 2) undergoing radical hysterectomy and bilateral pelvic lymph node dissection, while Group B was composed of 54 analogs receiving adjuvant radiotherapy following incident extrafascial hysterectomy. Treatment consisted of external beam radiotherapy (EBRT) and high-dose-rate intravaginal brachytherapy (HDRIVB). Initially, the whole pelvis was treated with 10 MV X-rays. After irradiation (44 Gy in 22 fractions over 4-5 weeks), the field was limited to the true pelvis and a further 10-20 Gy delivered in 5-10 fractions. For 21 patients in group A without pelvic lymph node metastasis or lymphovascular invasion, the radiation field was confined to the lower pelvis, with a prescribed dose of 50-58 Gy delivered over 5-6 weeks. HDRIVB was performed using an Ir-192 remote after-loading technique at 1-week intervals. A total of 159 patients (97%) received two insertions, while 5 had only one. The standard prescribed HDRIVB dose was 7.5 Gy to the vaginal surface. Logistic regression analysis was performed for assessment of the factors associated with NRRIII. RESULTS: After 38-119 months of follow-up (median, 60), 22 patients (13.4%) developed Radiation Therapy Oncology Group (RTOG) grade 2 or greater NRRIII at a median latency of 18 months (range, 5-48). Four patients were diagnosed as grade 3 complications requiring surgery and three had expired. The independent factors for NRRIII were radical hysterectomy (P = 0.04, relative risk 2.45), lower-pelvic dose >54 Gy (P = 0.0001, relative risk 10.27), and age over 60 years (P = 0.001, relative risk 5.45). The incidence of NRRIII for patients receiving whole and lower-pelvic irradiation was 14.5% and 10.6%, respectively (P = 0.45). Although there was no statistical significance comparing the two external beam irradiation strategies in terms of NRRIII, all four patients with grade 3 NRRIII underwent whole pelvic irradiation. CONCLUSION: This study identifies three predictive factors for the development of NRRIII following adjuvant radiotherapy for cervical cancer. Limiting the EBRT dose to less than 54 Gy, meticulous patient selection in the elderly, careful planning of the irradiated field, and the constraint of vaginal brachytherapy are four approaches to optimization of postoperative adjuvant radiotherapy.