Operative treatment of aortic dissections. Experience with 125 patients over a sixteen-year period.

An unselected, consecutive cohort of 125 patients underwent operative repair of acute and chronic aortic dissections with tubular graft interposition over a 16 year span. The absence of remote geographical referral biases and the unselected nature of this series provided a patient population that was representative of the disease process (as assessed heretofore only from autopsy series). Furthermore, this enabled high-risk subsets to be defined by retrospective analysis. Patients were classified according to whether the ascending aorta was involved (type A with involvement, type B without), irrespective of the site of intimal tear, and according to age of the dissection: Fifty-three patients had acute type A (Ac-A), 29 had chronic type A (Ch-A), 20 had acute type B (Ac-B), and 23 had chronic type B (Ch-B) dissections. Fourteen percent (17/125) of the dissections had ruptured. Concomitant aortic valve replacement (AVR) was performed in 11% (6/53) for Ac-A cases and 38% (11/29) of the Ch-A cases. A total of 391 patient-years of follow-up was analyzed; follow-up averaged 4.5 years and extended to 13.7 years. Over-all operative mortality rate was 34% (18/53) for Ac-A, 14% (4/29) for Ch-A, 45% (9/20) for Ac-B, and 22% (5/23) for Ch-B; during the most recent 5 year interval these figures were lower: 27%, 8%, 20%, and 20%, respectively, N = 50. Multiple preoperative variables were found to correlate significantly with both operative death and long-term survival. Operative survivors generally experienced satisfactory functional benefit. Late attrition averaged 8% per year; 61% of all late deaths were related to cardiac or cerebral causes. Over-all actuarial survival (+/- SEM) for the entire cohort was 54% +/- 5% at 5 years and 26% +/- 7% at 10 years; for the 89 patients surviving operation, these figures were 76% +/- 5% and 37% +/- 10%, respectively. No significant differences in long-term survival were evident between the different subgroups. Whether the primary intimal tear had been resected or concomitant AVR had been performed had no statistically significant bearing on operative mortality, functional result, necessity for late reoperation, or late attrition. The long-term "natural" history of surgically treated patients with aortic dissections, as defined in this study, should facilitate comparison with other treatment modalities. Results of the present analysis support immediate operative intervention for patients with Ac-A dissections and probably for those with Ac-B dissections. Additionally, surgical treatment of patients with symptomatic or enlarging Ch-A and Ch-B dissections provides satisfactory rehabilitation and long-term survival. Finally, we re-emphasize our recommendation for simplified classification of aortic dissections, based solely upon the presence or absence of ascending aortic involvement. Pathophysiology and expected biologic behavior pivot on this feature, and appropriate clinical strategy can thereby be defined.