Objective: The main objective was to assess the effectiveness of intraoperative cell salvage in terms of autotransfusion during cesarean section according to the different indications. The secondary objective was to identify a bleeding threshold that predicts autotransfusion. Design and population: This retrospective cohort study included every woman who received intraoperative cell salvage during a cesarean section at Lille University Hospital (France) between June 2007 and September 2018. Methods: The effectiveness of intraoperative cell salvage was characterized according to its ability to enable the reinfusion of autologous blood. The indications for intraoperative cell salvage were compared between those receiving versus those not receiving autotransfusion. Results: Among 293 patients who received intraoperative cell salvage, 97 (33.1%) were reinfused (95% confidence interval 27.7–38.8). Five had sickle cell disease. Placenta accreta, (P < .001), uterine fibroids (P = 0.013), and postoperative bleeding (P < .001) were independently associated with reinfusion; the respective odds ratios were 3.42 (1.8–6.54), 3.75 (1.49–9.38), and 10.15 (2.67–38.53). Using receiver-operating characteristic curve analysis, we defined 936 mL as the bleeding threshold to predict reinfusion with intraoperative cell salvage (positive predictive value = 0.77 and negative predictive value = 0.95). Conclusion: Placenta accreta, uterine fibroids, and postoperative bleeding seem to predict autologous blood transfusion when intraoperative cell salvage is used during cesarean section. In these contexts and when transfusion difficulties are expected, it is necessary to anticipate and set up the intraoperative cell salvage system to reduce the need for allogeneic blood transfusion in young women.