Antegrade selective cerebral perfusion in operations on the proximal thoracic aorta

Background. To determine the factors that influence hospital death and neurologic complications after surgery on the thoracic aorta using circulatory arrest and antegrade selective cerebral perfusion. Methods. From May 1989 through April 1997, 106 patients underwent surgery on the thoracic aorta using circulatory arrest and antegrade selective cerebral perfusion. Mean age was 63.0 +/- 11.5 years. Unilateral antegrade cerebral perfusion was used in 37 patients (35%), bihemispheric antegrade cerebral perfusion in 69 patients (65%). Mean antegrade cerebral perfusion time was 50.5 +/- 20.5 minutes. Indication for surgery was atherosclerotic aneurysm in 60 (56.5%) patients, postdissection aneurysm in 26 (24.4%), acute type A dissection in 16 (15.1%), other in 4 (4.0%). Results. Hospital mortality was 8.5% (n = 9; 70% CL: 5.8%-11.2%). Independent predictors of hospital mortality were rethoracotomy (odds ratio 5.7, p = 0.02), postoperative temporary (odds ratio 17.3, p = 0.02) or permanent (odds ratio 7.5, p = 0.03) neurologic dysfunction, postoperative dialysis (odds ratio 9.9, p = 0.008). Bilateral antegrade selective cerebral perfusion had a favorable impact on hospital mortality (odds ratio 0.08, p = 0.007). Temporary neurologic dysfunction occurred in 3.8% of patients (n = 4; 70% CL: 2.0%-5.6%); preoperative hemodynamic instability (odds ratio 14.8, p = 0.05) and perioperative technical problems (odds ratio 22.2, p = 0.033) were independent determinants of temporary neurologic dysfunction. Permanent central neurologic damage occurred in 5.4% of patients (n = 6; 70% CL: 3.2%-7.6%). Preoperative hemodynamic instability (odds ratio 18.9, p = 0.009) and approach through a left thoracotomy (odds ratio 9.4, p = 0.031) were significant predictors of permanent neurologic damage. Conclusions. Hospital mortality is affected significantly by the choice of technique used for antegrade cerebral perfusion. The incidence of both temporary and permanent postoperative central neurologic damage is influenced by preoperative hemodynamic instability. Duration of cerebral perfusion had no influence on the postoperative neurologic outcome. (C) 1999 by The Society of Thoracic Surgeons.