Learning from the last ultrasound. A population-based study of patients with abdominal aortic aneurysm.

Background: Patients with abdominal aortic aneurysm often are followed up with serial ultrasound examinations, but published studies usually describe rupture risk according to the diameter of the abdominal aortic aneurysm at diagnosis rather than by most recent ultrasound. Information in this form is misleading when used to predict prognosis.

Methods: We used data from the population-based cohort of residents of Rochester, Minn, diagnosed as having abdominal aortic aneurysm who have had at least 1 ultrasound measurement. Of the 181 patients who were enrolled in this cohort between January 1, 1974, and December 31, 1988, 5 had clinical evidence of rupture at entry. Analysis of a cohort defined by size category at "last ultrasound" was undertaken to assess rupture risk and growth rate.

Results: Median overall aneurysmal growth rate was 0.21 cm/y. Initial growth rate did not correlate with subsequent growth rate (r = 0.18; P = .14) or with initial size (r = -0.12; P = .22). Only 1 aneurysm ruptured when last ultrasound was less than 5 cm, and this occurred 3 1/2 years after this ultrasound. Estimated rupture risk by last ultrasound was 0% per year (95% confidence interval [CI], 0%-5%) when less than 4.00 cm, 1.0% per year (95% CI, 0%-5%) when 4.00 to 4.99 cm, and 11% per year (95% CI, 1%-21%) when 5.00 to 5.99 cm.

Conclusions: The most clinically useful approach to estimating the risk of abdominal aortic aneurysm rupture is according to size at last ultrasound. Aneurysm growth rate is predicted neither by size nor by initial growth rate.