Modeling of predissection aortic size in acute type A dissection: More than 90% fail to meet the guidelines for elective ascending replacement.

Objectives: The current guidelines for ascending aortic replacement were determined from already dissected aorta diameters. Previous computed tomography-based work on humans who underwent imaging before and directly after aortic dissection onset has shown an average 30% increase in the ascending aortic diameter with acute dissection. The present investigation evaluated the incidence of predissection ascending aortic dilatation in acute type A dissection.

Methods: From 2002 to 2013, 495 patients presented with acute type A aortic dissection to 1 center. Of these cases, 343 were non-Marfan, nonbicuspid with spontaneous dissection etiology. In those with available preoperative computed tomography angiograms (n = 83) or transesophageal echocardiograms (n = 260), the predissection ascending aorta diameters were modeled from the dissected aorta diameters by subtraction of the average diameter increase rate.

Results: Altogether 343 patients were included (age, 62 years; range, 53-73; 64% men). The median modeled predissection ascending diameter was 3.7 cm (first quartile, 3.3; third quartile, -4.1). Of the 343 patients, 334 (97%) and 315 (92%) had an ascending diameter before dissection onset of <5.5 cm and <5.0 cm, respectively. More than 60% of women and 70% of men had a nondilated ascending aorta before type A dissection onset. The median predissection ascending aortic diameter did not differ between women and men (3.7 cm; first quartile, 3.4; third quartile, 4.2; vs 3.7 cm; first quartile, 3.3; third quartile, 4.1; P = .810).

Conclusions: More than 60% of patients with spontaneous, non-Marfan, nonbicuspid, type A dissection will have a nondilated ascending aorta before dissection onset. Only 3% would meet the criteria for elective ascending replacement to prevent aortic dissection. Additional research on the genetic and biochemical predictors of aortic dissection is essential.