The purpose of this study was to quantitatively analyze normal and preeclamptic uteroplacental vasculature. Myometrial arteries from eight placental bed biopsies from uncomplicated term deliveries and 12 from proteinuric preeclampsia were characterized as uteroplacental, spiral, or basal arteries. Basal lumens within 0.2 mm radius and spiral/uteroplacental lumens within 0.4 mm radius were considered as the same artery. The biopsy area, lumen density, and arterial density (after correction for multiple lumens), lumen area, lumen perimeter, mean wall thickness, inflated diameter, and a slant factor, measuring the obliqueness of arterial transection, and ratios of lumen characteristics to mean wall thickness were analyzed. In preeclamptic cases, there were more basal lumens/mm2 and basal arteries/mm2 (P=.003, P=.03), and more spiral lumens/mm2 and spiral arteries/mm2 (P = .01, P = .03). Basal lumen area (P = .0003) and wall thickness (P = .007), and basal and spiral artery lumen perimeters and inflated diameters (for each, P = .0001, P = .048, respectively) and inflated diameter/wall ratios (P = .04, P = .05) were reduced compared with normal cases. Preeclamptic spiral and basal arteries are more tortuous or densely distributed than normal placental bed arteries, with smaller-caliber lumens and thicker walls. Failure of proper placentation may result in abnormal spatial anatomy in the placental bed. Alternatively, an anatomic variant of spiral and basal arteries may be more susceptible to hemodynamic stresses and endothelial damage and may predispose to preeclampsia.