CorrespondenceDoes the presence of a uterine scar influence the site of placental implantation? Naji et al. 1 compared the incidence of low-lying placenta in 738 patients who had one previous Cesarean section (CS) (594 women) or more than one previous CS (144 women) with 1856 controls in a case-control study. A low-lying placenta overlapping the CS scar increases dramatically the risk of placenta accreta with its major complications 2 , and prospective studies investigating placental location can make considerable contributions to the appropriate management of patients with one or more previous CS. Naji et al. did not find a significant difference in the incidence of low-lying placenta in patients with one or more previous CS compared with controls, although they concluded that the presence of a uterine scar did influence placentation, with more posterior and fewer fundal placentae in the CS group.Actually, several population-based studies 2,3 have demonstrated that women whose first delivery was by CS have a significantly increased risk of placenta previa at their second delivery compared with women with prior vaginal deliveries, with an increase in relative risk of 1.5 2,3 , and in those whose first two deliveries were by CS, the relative risk is higher still (2.0; 95% CI, 1.3-3.0) 3 compared with that in women whose first two deliveries were vaginal. The population of the CS group in Naji et al.'s study was probably not sufficiently powered to detect this association.Naji et al. 1 suggested diverse mechanisms that could explain abnormal placentation after CS, specifically disruption of the junctional zone and impaired decidualization. Jauniaux et al. 4 recently supported this concept of a primary iatrogenic deciduomyometrium defect in placenta accreta, with failure of decidualization in the area of a previous uterine scar, which exposes the myometrium to trophoblastic invasion and subsequent excessive vascular remodeling of the radial and arcuate arteries below the junctional zone. However, while these mechanisms may explain how the placentation is abnormal in depth and vasculature, they do not explain abnormal location. Naji et al. found that women with a low-lying placenta were significantly older than were women with a fundal placenta. The risk of placenta accreta is particularly high when the trophoblast overlaps the scar, and maternal age, which correlates both with low-lying placenta 1,5 and with previous CS 2,3 , could be the link. The population of the CS group in the study of Naji et al. was probably not sufficiently powered to detect this association either. Miller et al. 2 also showed that among women with placenta previa, advanced maternal age (≥35 years) and previous CS delivery are independent risk factors for placenta previaaccreta, which is a major and increasing cause of heavy postpartum hemorrhage 2 .L. Carbillon