This study was designed to develop a preoperative predictor model using multidetector CT (MDCT) imaging findings for the prediction of surgical outcome in patients with advanced epithelial ovarian cancer (EOC). Seventy-seven patients with advanced EOC who had preoperative MDCT and who had undergone primary cytoreductive surgery between January 1999 and June 2008 were enrolled in the study. All MDCT examinations were analysed retrospectively, assessing nine imaging features without knowledge of the operative findings. The complete debulking rate and optimal debulking rate were 42.5% and 93.2%, respectively. Based on the use of univariate and multivariate analysis, one imaging feature that showed the presence of upper abdominal ascites was significantly associated with incomplete tumour resection (P = 0.019 on univariate analysis, P = 0.04 on multivariate analysis). The combination of the MDCT findings that showed the presence of upper abdominal ascites and diffuse subdiaphragmatic peritoneal nodularity was associated with an incomplete resection of a tumour (P = 0.04; specificity 80.0%; positive predictive value 76.9%). It is important that emphasis on the upper abdominal sites of the tumour is helpful when reporting preoperative MDCT results.