J. L. Philip1, B. Rademacher1, C. B. Goodavish2, P. D. DiMusto3, N. C. De Oliveira2, P. C. Tang2 1University Of Wisconsin,General Surgery,Madison, WI, USA 2University Of Wisconsin,Cardiothoracic Surgery,Madison, WI, USA 3University Of Wisconsin,Vascular Surgery,Madison, WI, USA
Introduction: Acute type A aortic dissection is a surgical emergency. Traditional classifications systems group dissections were based on an evolving therapeutic approach. Using statistical groupings based on dissection morphology, we examined the impact of morphology on presentation and clinical outcomes.
Methods: We retrospectively reviewed 108 patients who underwent acute type A dissection repair from 2000-2016. Dissection morphology was characterized using 3-dimensional reconstructions of computed tomography scan images based on the true lumen area as a fraction of the total aortic area along the aorta. Two-step cluster analysis was performed to group the dissections.
Results: Cluster analysis resulted in two distinct clusters (silhouette cluster of cohesion and separation = 0.6). Cluster 1 (n=71, 65.7%) was characterized by a dissection extending form the ascending aorta into the abdominal aorta and iliac arteries (table 1). Cluster 2 dissections (n=37, 34.3%) extends from the ascending aorta to the aortic arch with limited extension into the distal arch and descending thoracic aorta. Cluster 2 extends the traditional DeBakey type II definition to include dissections propagating into the arch. Cluster 1 patients had more malperfusion (P=0.002) as well as lower extremity and abdominal pain on presentation (P<0.05). Cluster 2 had a greater number of diseased coronary vessels (P<0.05) and more commonly had previous percutaneous coronary intervention (P<0.05). No differences in age, gender and other major comorbidities were noted. Cluster 1 is characterized by a smaller primary tear area (3.7 vs 6.6 cm2, P=0.009), greater number of secondary tears (1.9 vs 0.5, P<0.001), more dissected non-coronary branches (2.9 vs 0.6, P<0.001) and greater degree of aortic valve insufficiency (P<0.05). Operative variables including cardiopulmonary bypass and cross-clamp time as well as extent of arch repair and type of proximal operations were similar. There were no differences in post-operative complications or survival. Cluster 1 had a significantly higher rate of intervention for distal dissection complications (10% vs 0%, P=0.048).
Conclusions: This study examines clinical presentation and outcomes in acute type A dissection based on morphology using statistical categorization. Cluster 2 acute type A dissections had much less distal aortic dissection involvement and need for distal aortic intervention. Therefore, it is likely reasonable to extend the definition of DeBakey type II dissection to involvement of the distal arch and proximal descending thoracic aorta. The greater area of the primary aortic tear in Cluster 2 with rapid decompression of the false lumen may explain the lesser degree of distal aortic dissection.
