Aim:: To identify parturients at risk of inability to extend labor epidural analgesia in whom alternative methods of anesthesia should be considered for Cesarean section (CS). Methods:: For 6 months, we prospectively studied women undergoing CS with a functioning epidural catheter in place from the delivery ward. All parturients received the same epidural protocol: bolus of bupivacaine 0.1% and fentanyl, followed by bupivacaine 0.1% and fentanyl (2 g/ml) 10–15 ml/h and an additional 5 ml of bupivacaine 0.125% as top-up according to patient request. Sixteen milliliters of lidocaine 2%, 1 ml of bicarbonate and 100 g of fentanyl were given for CS. Failed epidural analgesia was defined as the need to convert to general anesthesia. Results:: Of the 101 parturients studied, 20 (19.8%) required conversion to general anesthesia. In univariate analysis, the likelihood of failed epidural anesthesia was inversely correlated with parturient age (P = 0.014) and directly correlated with pre-pregnancy weight (P = 0.019), weight at the end of pregnancy (P = 0.003), body mass index at the end of pregnancy (P = 0.0004), gestational week (P = 0.008), number of top-ups (P = 0.0004) and visual analog scale (VAS) score 2 h before CS (P = 0.03). In multivariate analysis, the number of top-ups in the delivery ward was the best predictor of epidural anesthesia failure (odds ratio, 4.39; P = 0.005). Conclusion: Younger, more obese parturients at a higher gestational week, requiring more top-ups during labor and having a higher VAS score in the 2 h before CS are at risk for inability to extend labor epidural analgesia to epidural anesthesia for CS.