Purpose: On the basis of a prospective analysis of 147 patients undergoing surgery for ruptured abdominal aortic aneurysm (AAA) and recorded in the Canadian Society for Vascular Surgery Aneurysm Registry, this study defines the early and 6-year actuarial survival rates and determines the predictive variables that are associated with survival.
Methods: Ongoing follow-up of a cohort of patients was current at the time of analysis. To identify the preoperative, intraoperative, and postoperative variables that were associated with survival, statistical methods included chi-squared analysis, logistic regression analysis, Kaplan-Meier analysis, and Cox regression analysis.
Results: The survival rate was 48.6% at 1 month, 34.7% +/- 4.2% at 3 years, and 22.0% +/- 4.0% at 6 years. When preoperative and intraoperative variables were considered and logistic regression analysis was used, the highest probability of early in-hospital survival was associated with preoperative creatinine levels of 1.3 mg/dl or less, intraoperative urine output of 200 ml or greater, and infrarenal clamp site. The highest probability of late survival, as calculated by the Cox proportional hazards method, was predicted by the patient's age and total urine output during the procedure. When all variables, including postoperative complications, were considered, late survival was highest if intraoperative urine output was 200 ml or greater and respiratory failure and myocardial infarction did not occur. For those patients with ruptured AAA who survived operation (i.e., greater than 1 month), the long-term survival rate was significantly lower than a comparable group undergoing repair of nonruptured AAA.
Conclusions: Patients who survive repair of a ruptured AAA have a lower late survival rate than patients undergoing elective repair. When a patient is evaluated before operation, no combination of preoperative variables could identify those patients with little or no chance of survival; hence, the decision to repair a ruptured AAA should be made on clinical grounds. However, after surgery (when information on intraoperative and postoperative variables is also available), the results of this study provide a basis for the surgeon to use these prognostic variables to assist clinical judgment and guide discussions on prognosis with the family and to identify those patients who have such a low chance of early and late survival that further aggressive treatment may be futile.