Actuarial analysis of variables associated with rupture of small abdominal aortic aneurysms.

This study identified risk factors associated with rupture of small abdominal aortic aneurysms in patients initially selected for nonoperative management. Sixty-seven patients, 53 men and 14 women, 50 to 91 years of age (mean 72 years) were followed 3 to 99 months (mean 36 months). All patients underwent serial aortic ultrasonography. The annual rate of aneurysm rupture was 6%, with an annual mortality rate caused by rupture of 5% and an annual mortality rate caused by coexistent disease of 6%. Thirty potential risk factors, including blood pressure, aneurysm size measured by ultrasonography, rate of aneurysm expansion, smoking, serum cholesterol levels, and cardiac, pulmonary, and renal risks, were analyzed by Cox proportional hazards regression to identify variables related to rupture. Aneurysm anteroposterior expansion rates varied from 0 to 1.5 cm/year but were not different in aneurysms that ruptured. Only diastolic blood pressure, initial aneurysm anteroposterior diameter, and degree of obstructive pulmonary disease were independently predictive of rupture (p less than 0.05, Wald test). With these data, actuarial rupture rates were predicted for patients with selected values of these three covariates. Predicted 5-year rupture rates varied from 2% when these risk factors were absent to 100% when all three risk factors were significant. Obstructive pulmonary disease, initial aneurysm size, and diastolic hypertension must be evaluated prospectively to assess their accuracy in predicting small aneurysm rupture.