OBJECTIVE To determine the 30-day morbidity and mortality rates for patients with an intact uterus undergoing high-dose-rate (HDR) brachytherapy, and to assess risk factors which may predict for these potentially life-threatening complications. METHODS From August 1989 to December 1994, 128 cervical and 41 medically inoperable endometrial cancer patients were treated with 5 outpatient weekly HDR brachytherapy insertions. Patients with cervical cancer also were treated with external beam radiotherapy. Acute events that resulted in either hospitalization (morbidity) or death (mortality) within 30 days of the implant were analyzed. Univariate and multivariate analyses were performed to identify risk factors. RESULTS Overall there were 16 acute events in 169 patients (9.5%). The overall morbidity and mortality rates for the cervical and endometrial patients were 5.5%, 1.6%, 7.3%, and 9.8%, respectively. The following factors were significant by univariate analysis: age per decade, American Society of Anesthesiologists (ASA) score, Karnofsky Performance Status (KPS), significant medical history, diagnosis of cervical vs. endometrial cancer, and mean time exceeding 160 minutes for the procedure. Since age was the most significant predictive factor (p = 0.0003), bivariate analyses were performed by adjusting for age. In these analyses only ASA and KPS maintained significance, while a positive medical history was of borderline significance (p = 0.07). CONCLUSIONS The morbidity and mortality rates observed in gynecologic patients selected for HDR brachytherapy were similar to low-dose-rate, but higher than other HDR reports. Reasons for this include a higher risk population, especially those with medically inoperable endometrial cancer. In the cervical cancer patients, some of the complications may have also been a result of the external beam portion of the radiation. In order to minimize the acute complications observed in the present HDR brachytherapy system, the following changes have been implemented: appropriate patient selection, anesthesiology involvement to monitor conscious sedation for high-risk patients, external beam radiotherapy alone in patients at extremely high risk, deep vein thrombosis (DVT) prophylaxis, use of intraoperative ultrasound, shorter duration in the brachytherapy suite, and preradiation treatment plans (plans executed prior to the insertion) if applicable. Finally, this analysis suggests that these procedures should be performed in a hospital-based setting where appropriate support is available.