To estimate maternal outcome after conservative management of placenta accreta. This retrospective multicenter study sought to include all women treated conservatively for placenta accreta in tertiary university hospital centers in France from 1993 to 2007. Conservative management was defined by the obstetrician's decision to leave the placenta in situ, partially or totally, with no attempt to remove it forcibly. The primary outcome was success of conservative treatment, defined by uterine preservation. The secondary outcome was a composite measure of severe maternal morbidity including sepsis, septic shock, peritonitis, uterine necrosis, fistula, injury to adjacent organs, acute pulmonary edema, acute renal failure, deep vein thrombophlebitis or pulmonary embolism, or death. Of the 40 university hospitals that agreed to participate in this study, 25 institutions had used conservative treatment at least once (range 1-46) and had treated a total of 167 women. Conservative treatment was successful for 131 of the women (78.4%, 95% confidence interval [CI] 71.4-84.4%); of the remaining 36 women, 18 had primary hysterectomy and 18 had delayed hysterectomy (10.8% each, 95% CI 6.5-16.5%). Severe maternal morbidity occurred in 10 cases (6.0%, 95% CI 2.9-10.7%). One woman died of myelosuppression and nephrotoxicity related to intraumbilical methotrexate administration. Spontaneous placental resorption occurred in 87 of 116 cases (75.0%, 95% CI 66.1-82.6%), with a median delay from delivery of 13.5 weeks (range 4-60 weeks). Conservative treatment for placenta accreta can help women avoid hysterectomy and involves a low rate of severe maternal morbidity in centers with adequate equipment and resources.