Patients with small recurrent cervical carcinomas following radiation therapy may be salvaged with radical hysterectomy rather than exenteration. Between 1953 and 1993, 50 patients underwent radical hysterectomy for persistent (n = 18) or recurrent (n = 32) cervical cancer after primary radiotherapy. The mean age of the cohort was 44 years (range, 23-70). Histologic types were squamous in 46, adenocarcinoma in 3, and adenosquamous in 1. Of 37 patients with staged disease, 24 had stage IB/IIA, 7 had stage IIB, 2 had stage IIIA, and 2 had stage IIIB. Combination radiotherapy, consisting of 40-45 Gy external-beam radiation plus brachytherapy (mean 6980 mg/hr), was performed in 32 patients (64%). In the 32 patients with recurrent lesions, the median interval from definitive radiotherapy to radical hysterectomy was 16 months (4-301), with 19 of these patients (60%) presenting within the first 24 months. Patients with persistent carcinomas underwent radical hysterectomy after a median observation interval of 2 months (1-4). A class II or III radical hysterectomy was performed in 39 (78%) cases. Pelvic and para-aortic lymph node samplings were performed in 39 patients (78%), including 33 (66%) who underwent complete pelvic lymphadenectomy. Among those sampled, 5 (13%) had metastatic nodal disease. All 5 patients died of disease at a median 13 months after surgery. Severe postoperative complications occurred in 21 patients (42%). The most common site of injury was the urinary tract, with 14 patients (28%) developing vesicovaginal or rectovaginal fistulae, 11 (22%) developing ureteral injuries, and 10 (20%) developing severe long-term bladder dysfunction. There was one postoperative death from sepsis among the entire population. Patients with abnormal preoperative intravenous pyelograms (P < 0.05), patients with recurrent presurgical lesions (P < 0.05), and patients with postoperative pelvic cellulitis (P < 0.01) were more likely to develop fistulae. The 5- and 10-year actuarial survival rates for all cases was 72 and 60%, respectively. Tumor size at radical hysterectomy was significantly associated with survival. Five-year actuarial survival in 12 of 44 patients (27%) with identifiable lesion diameters less than 2 cm was 90% compared with 64% in patients with larger lesions (P < 0.01). Prolonged disease-free survival occurred in 26 of 50 patients (52%) who had known disease status at follow-up, whereas recurrence after radical hysterectomy was seen in 24 patients (48%).(ABSTRACT TRUNCATED AT 400 WORDS)