Background. We examined potential maternal-fetal interactions during fetal cardiopulmonary bypass. these interactions, not previously described, may significantly influence attempts at fetal cardiac surgery. Methods. Eight fetal sheep underwent cardiopulmonary bypass (5 singletons, 3 twins; 100-109 days) for 60 minutes using a centrifugal microcircuit (20 mL prime), and the placenta as oxygenator. We measured maternal hemodynamics, arterial blood gases, and changes in blood flow to the gravid uterus using bilateral uterine artery flow probes. Maternal measurements were correlated to fetal hemodynamics, blood gases, and umbilical blood flows. After bypass, fetuses were followed for 60 minutes. Results. Decreases in uterine blood flow occurred without changes in maternal hemodynamics or arterial blood gases, but were associated with worsening fetal arterial blood gases (pH decreased from 7.2 ± 0.2 to 7.0 ± 0.1, partial pressure of carbon dioxide increased 45.6% and partial pressure of oxygen decreased 15.4%). Changes in maternal hemodynamics (decreased systolic blood pressure [17.5%, SD = 11] and decreased diastolic blood pressure [20.3%, SD = 15]) were only noted when uterine blood flows decreased by greater than 38.2% (SD = 26). Correction of maternal hypocalcemia (0.89g/dL, SD = 0.1) led to improved uterine artery flows (28.3% increase, SD = 30). Finally, fetal sternotomy, cannulation, and cardiopulmonary bypass each decreased uterine artery flows by 27.5% (SD = 18), 31.0% (SD = 26), and 39.7% (SD = 25), respectively. Similar changes were not observed in the nonbypass twin. Conclusions. Significant changes in uterine blood flow can occur during fetal cardiopulmonary bypass support without apparent changes in maternal hemodynamics or arterial blood gases. These changes imply a unique transplacental maternal-fetal interaction. Limited data from the twin fetus suggest a localized mechanism involving only the segment of placenta exposed to extracorporeal circulation.