Factors influencing outcome after emergency surgical repair of acute type A aortic dissection.

OBJECTIVE: We retrospectively reviewed our more recent experience with acute type A aortic dissection in order to identify possible risk factors influencing current surgical results.

METHODS: Between January 1990 and January 1998, 122 patients (86 males and 36 females; mean age 60 +/- 12 years) underwent emergency repair of acute type A aortic dissection using a standard surgical approach. Seventy-four (61%) patients required isolated replacement of the dissected ascending aorta, 27 (22%) required additional replacement of the aortic arch and 21 (17%) required total aortic root replacement. Surgical outcome was evaluated in terms of operative mortality and morbidity. Results of patients presenting with preoperative complications (Group C) (i.e. cardiac tamponade, cerebral stroke, cardiogenic shock, acute myocardial infarction, anuria or visceral ischemia) were compared with those of uncomplications cases (Group U) and with a calculated risk of expected operative mortality (EOM-rate) based on an analysis of each patient set of preoperative risk factors. Sixteen preoperative and 18 perioperative variables were also analyzed to identify conditions influencing morbidity and mortality.

RESULTS: Fifty-seven patients (47%) presented with preoperative complications (Group C) and 65 (53%) did not (Group U). Overall operative mortality was 22% (27 patients). Mortality within subgroups was 40 and 6% for complicated and uncomplications cases, respectively (p < 0.001). The 85% of the overall mortality occurred in Group C patients. During the experience, the operative mortality rate actually observed ranged from 0 to 38% and was similar to the calculated expected risk, thus proving a direct relationship with the amount of complicated cases operated on each year. Multivariate analysis revealed that older age and hemopericardium significantly increased the risk of operative death, while male gender, preoperative complications, postoperative bleeding, duration of circulatory arrest and aortic cross-clamp time significantly predicted morbidity (p = 0.02).

CONCLUSIONS: Current results of emergency repair of acute type A aortic dissection are strictly dependent on the number of complicated cases referred for operation. Earlier diagnosis and prompt referral before development of preoperative complications appear essential to improve surgical results.