Assessment of hepatosplanchnic pathophysiology during thoracoabdominal aortic aneurysm repair using visceral perfusion and shunt

Objective: Despite the recognition of importance to avoid visceral ischemia during thoracoabdominal aortic aneurysm (TAAA) repair, the methodology of visceral perfusion seems still controversial and its pathophysiology has not been clearly understood. We investigated hepatosplanchnic metabolism during visceral perfusion/shunt in TAAA repair. Methods: Seventeen patients (10 male, 64 +/- 15 years old) who underwent elective TAAA repair using visceral perfusion/shunt under mild hypothermic distal aortic perfusion were retrospectively enrolled. Their aneurysm extension was type I and II in eight patients. In seven patients, four visceral arteries were perfused through a side-arm of distal aortic perfusion, white they were perfused by an independent pump in another five patients. In four of these 12 (two in each technique), visceral perfusion was converted into selective shunt after completion of aortic anastomosis. In the remaining five patients, four branches were initially perfused through a side-arm of distal aortic perfusion, and aortic perfusion was subsequently stopped after completion of aortic anastomosis. Hepatic venous oxygen saturation (ShO(2)), oxygen and lactate extraction ratio (OER, LER), and arterial ketone body ratio (AKBR) were measured at six time points. Results: There was no mortality, liver/renal dysfunction, or spinal cord injury. Two patients required re-exploration for bleeding. Fourteen patients were extubated within 24 h postoperatively. Mean intensive care unit stay was 2.3 +/- 1.7 days. During visceral perfusion, OER raised (31 +/- 13% to 68 +/- 21%, p = 0.0012) and ShO(2) decreased (67 +/- 12% to 34 +/- 24%, p = 0.0026) significantly. They recovered to baseline at skin closure. During the same period, LER (41 +/- 22% to -1 +/- 34%, p = 0.0035) and AKBR (0.47 +/- 0.13 to 0.20 +/- 0.08, p = 0.0012) significantly decreased. AKBR recovered to baseline at skin closure, but LER did not. ShO(2) (R-2 = 0.483, p = 0.0257) and LER (R-2 = 0.774, p = 0.0018) at skin closure and LER after initiation of partial cardiopulmonary bypass (R-2 = 0.427, p = 0.0211) had significant correlation with postoperative peak serum bilirubin level. AKBR after initiation of partial cardiopulmonary bypass had significant correlation with postoperative peak serum alanine aminotransferase level (R-2 = 0.289, p = 0.0476). Conclusions: Visceral perfusion/shunt in TAAA repair may avoid critical irreversible hepatosplanchnic ischemia but provide unphysiological blood flow to the liver and thus should be shortened. (C) 2009 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.