Impact of Different Cannulation Strategies on In-Hospital Outcomes of Aortic Arch Surgery: A Propensity-Score Analysis
BACKGROUND:
The impact of different cannulation strategies on outcomes of aortic arch surgery remains controversial. This retrospective study sought to evaluate central cannulation (ascending aorta, right axillary, and innominate artery) compared with femoral artery cannulation for aortic arch surgery, and to identify among preoperative and intraoperative variables the independent predictors of death and permanent neurologic dysfunction (PND) in aortic arch surgery.
METHODS:
All patients were operated through a median sternotomy using antegrade selective cerebral perfusion with moderate hypothermia as a method of brain protection. Treatment bias was addressed by use of propensity-score matching and multivariate regression analysis. Logistic regression models were used to identify the independent predictors of hospital mortality and PND.
RESULTS:
Of the 473 patients undergoing aortic arch surgery, 273 (57.7%) underwent femoral cannulation (FC), and 200 (42.3%) underwent central cannulation (CC). The CC and FC cannulation were associated with similar risk of in-hospital death (absolute risk reduction [ARR]: 0.7%; p = 0.880) and PND (ARR:-2.6%, p = 0.361) in the overall cohort and after adjusting for propensity-based matching (ARR for hospital mortality: 2.2%, p = 0.589; ARR for PND: 3.4%, p = 0.271). Female gender (odds ratio [OR]:2.1, p = 0.030), type A acute dissection or intramural hematoma (OR: 2.2; p = 0.041), and CPB time (OR: 1.010/minute, p = 0.015) were independent predictors of in-hospital death. Female gender (OR: 2.4; p = 0.033), type A acute dissection or intramural hematoma (OR: 4.2; p = 0.005), and diabetes (OR: 6.6, p = 0.007) were independent predictors of PND.
CONCLUSIONS:
During aortic arch surgery, CC and FC are associated with a similar risk of postoperative death and PND. Type A acute aortic dissection and cardiopulmonary bypass time remain strong risk factors for mortality and PND.
