Retrograde cerebral perfusion provides negligible flow through brain capillaries in the pig

Objectives: Although retrograde cerebral perfusion is being used clinically during aortic arch surgery, whether retrograde flow perfuses the brain eff ectively is still uncertain. Methods: Fourteen pigs were cooled to 20 degreesC with cardiopulmonary bypa ss and perfused retrogradely via the superior vena cava for 30 minutes: 7 u nderwent standard retrograde cerebral perfusion and 7 underwent retrograde perfusion with occlusion of the inferior vena cava. Antegrade and retrograd e cerebral blood flow were calculated by quantitating fluorescent microsphe res trapped in brain tissue after the animals were put to death; microspher es returning to the aortic arch, the inferior vena cava, and the descending aorta were also analyzed during retrograde cerebral perfusion. Results: Antegrade cerebral blood flow was 16 +/- 7.7 mL . min(-1) . 100 g( -1) before retrograde cerebral perfusion and 22 +/- 6.3 mL . min(-1) . 100 g(-1) before perfusion with caval occlusion (P =.14). During retrograde per fusion, calculations based on the number of microspheres trapped in the bra in showed negligible flows (0.02 +/- 0.02 mL . min(-1) . . 100 g(-1) with r etrograde cerebral perfusion and 0.04 +/- 0.02 mL . min(-1) . 100 g(-1) wit h perfusion with caval occlusion; P =.09): only 0.01% and 0.02% of superior vena caval inflow, respectively. Less than 13% of retrograde superior vena caval inflow blood returned to the aortic arch with either technique. Duri ng, retrograde cerebral perfusion, more than 90% of superior vena caval inp ut was shunted to the inferior vena cava and was then recirculated, as indi cated by rapid development of an equilibrium in microspheres between the su perior and inferior venae cavae. With retrograde perfusion and inferior ven a caval occlusion, less than 12% of inflow returned to the descending aorta and only 0.01% of microspheres. Conclusions: The paucity of microspheres trapped within the brain indicates that retrograde cerebral perfusion, either alone or combined with inferior vena caval occlusion, does not provide sufficient cerebral capillary perfu sion to confer any metabolic benefit. The slightly improved outcome previou sly reported with retrograde cerebral perfusion during prolonged circulator y arrest in this model may be a consequence of enhanced cooling resulting f rom perfusion of nonbrain capillaries and from venoarterial and venovenous shunting.