Introduction: Pregnancy in sickle cell womenis associated with adverse maternal outcomes. Studies evaluating the effects of the transfusion program (TP) during pregnancy are contradictory. The objective of this retrospective study was to evaluate the effect of the transfusion program on maternal and neonatal outcomes in Sickle cell disease (SCD) . Since the publications highlighting the frequency of delayed hemolysis transfusion reaction (DHTR) in SCD patients and particularly in pregnant women, transfusion(TF) indications have changed in the management of pregnancies. Transfusion indications have been limited since 2015 in view of the risks of alloimmunization and DHTR, associated with the lack of evidence of obstetrical benefit, in SCD pregnancies (10). Our objective was to evaluate the impact of the transfusion program during pregnancy on obstetric and neonatal outcomes with a propensity score to account for potential biases. Method: This is a retrospective cohort study based on analysis of electronic and paper medical records including all pregnancies in women with SCD (Hb SS, SC, S beta-thalassemia phenotypes) who received prenatal care delivered in a French tertiary care academic center (CHI-Créteil) and in Henri Mondor SCD referral center between January 2004 and December 2017. The DHTR risk score was defined for all patients. Confounding factors considered in a propensity score were delivery period, SCD genotype, history of acute chest syndrome (ACS), and delayed hemolytic transfusion reaction (DHTR) risk score. Results: We analyzed 246 pregnancies in 173 patients. Twenty-two pregnancies with a history of DHTR were analyzed separately. We found a higher frequency of TP before 2013 (119/148 [80.4%] vs 38/76 [50%], p<0.001). Patients with a history of ACS and a low DHTR risk score were significantly more represented in the transfusion program group. There was no difference between the two groups in the frequency of preeclampsia, eclampsia, Hellp syndrome, term of delivery , births <37 weeks, birth weight, and cesarean section. Punctual transfusions were more frequent in the group without TF program. Rates of preterm birth before 34 weeks, vaso-occlusives crisis(VOC) and ACS during pregnancy were significantly more frequent in the group without a transfusion program. There were 4 inaugural DHTRs in each of the 2 groups and 4 deaths in the TF Program group; this difference was not significant. Multivariate analysis weighted by propensity score showed that TF program was associated only with a significant reduction in the risk of preterm delivery before 34 weeks (OR= 0.05 95% CI [0.01-0.31], p = 0.001). Of the 22 pregnancies with a history of DHTR, 4 were complicated by a new DHTR event after emergency transfusion. The rates of delivery before 34 and 37 Weeks of gestation, neonatal death, frequency of ACS, hospitalization during pregnancy and in intensive care unit, and composite adverse obstetric outcome were significantly different between the group with a history of DHTR and all patients without known DHTR, confirming the potential seriousness of these patients. Conclusion: This study shows a trend toward reduced routine use of the transfusion program during pregnancy. There was no difference between the two groups( TP and without TP) in the frequency of pre-eclampsia, eclampsia, Hellp syndrome, term of delivery and births <37 weeks, birth weight and caesarean section. Propensity score-weighted analysis reveals an independent protective effect of the transfusion program on prematurity before 34 weeks as well as composite obstetric criteria (birth <34 weeks and/or preeclampsia and/or SGA and/or abruption and/or IUFD and/or maternal death and/or neonatal death). Although the benefit/risk balance of the transfusion program should be considered on a case-by-case basis with the DHTR risk score in early pregnancy. In patients with low transfusion risk, transfusions may be recommended upon discontinuation of HU or in case of VOC complications. Prospective studies are still needed on large homogeneous patient populations and the evaluation of alternative treatments to transfusion such as hydroxycarbamide seems important.