OBJECTIVE: To determine whether the increased risk of placenta previa at delivery in patients with a prior cesarean results from an increased risk of abnormal implantation or a lower likelihood of resolution. METHODS: A hospital-based, case-control study was performed. Cases were defined as singleton pregnancies with a placenta previa on second-trimester ultrasonography. Controls, chosen randomly from patients without a placenta previa on second-trimester ultrasonography, were matched 3:1 with cases. Odds ratios (OR) were calculated, controlling for other independent risk factors for previa: age, parity, race, and smoking. RESULTS: There were 88 cases identified and 264 controls. Twenty cases (22.7%) and 35 controls (13.3%) had a history of prior cesarean delivery. Previous cesarean delivery was an independent risk factor for previa on second-trimester ultrasonography (OR 1.92, 95% confidence interval [CI] 1.04-3.55), as was the number of cesareans (OR 1.62, 95% CI 1.12-2.34). However, neither retained their significance after adjusting for other known risk factors for previa (OR 1.50, 95% CI 0.77-2.92, and OR 1.40, 95% CI 0.93-2.10, respectively). At delivery, a history of cesarean was associated with a 3-fold increased risk of previa. CONCLUSION: A previous cesarean delivery did not increase the odds for detecting a placenta previa on second-trimester ultrasonography. At delivery, a previous cesarean was associated with a previa, suggesting a lower likelihood of resolution.