Objective: To determine whether regional hypoperfusion of the interventricular septum occurs during ventilation with positive end-expiratory pressure. Design: Animal study. Animals: Anesthetized, closed chest dogs (n = 8). Interventions: Induction of experimental adult respiratory distress syndrome (ARDS) and then ventilation with 10,15, and 20 cm H2O of positive end-expiratory pressure. Measurements and Main Results: Cardiac output and regional interventricular septum blood flow 'were assessed at control, at induction of experimental ARDS, and at each level of positive end-expiratory pressure. Ventilation with 20 cm H2O of positive end-expiratory pressure decreased cardiac output (-32% vs. control, p <.05), and did not change absolute, but increased relative (to cardiac output) interventricular septum blood flow. During experimental ARDS and ventilation at 20 cm H2O end-expiratory pressure, there was a redistribution of flow toward the right ventricular free wall (+93%, p < .001) and the right ventricular part of the interventricular septum (+68%, p < .01), while flow to the left ventricular interventricular septum and to the left ventricular free wall remained unchanged. Locally hypoperfused interventricular septum areas or findings indicative of interventricular septum ischemia were not observed during positive end-expiratory pressure. Conclusions: The decrease in cardiac output during positive end-expiratory pressure is not caused by impaired interventricular septum blood supply. The preferential perfusion of the right ventricular interventricular septum indicates increased local right ventricular interventricular septum oxygen-demand and suggests that during positive end-expiratory pressure, this part of the interventricular septum functionally dissociates from the left ventricular interventricular septum and the left ventricular free wall to support the stressed right ventricle.