Abstract Objective To examine the clinical results of a preoperative image-guided pulse-dose-rate brachytherapy (PDR-BT) in early stage cervical cancer. Materials/methods We examined the outcome of consecutive patients with early stage cervical cancer undergoing preoperative image-guided PDR-BT between 2004 and 2013 because of risk factors (lymphovascular embols and/or tumour > 2 cm). The objective was to deliver 60 Gy to 100% of the intermediate risk clinical target volume. Brachytherapy was followed, 6–8 weeks later, by a radical hysterectomy/bilateral salpingo-oophorectomy plus pelvic +/− para-aortic lymph node dissection. Patients with positive lymph nodes had postoperative chemoradiation. Results 77 patients met the above criteria of preoperative PDR-BT. On hysterectomy specimen, 54 (70.1%) presented a complete histological response. Four (5.2%) had a tumour residuum ⩾ 1 cm. Median follow-up was 46.8 months. 5-Year disease-free survival (DFS) rate was 84.4%. Only one local recurrence was observed. The presence of lymph nodal metastases, a tumour size > 3 cm and a brachytherapy/surgery time interval ⩾ 9 weeks correlated with a poorer DFS. Six postoperative complications were encountered (7.8%). Total reference air kerma correlated with late vaginal toxicity ( p = 0.02). Conclusions A preoperative image-guided PDR-BT was safe and effective. Predictive factors for survival and toxicity were evidenced.