The incidence and mortality rate of ruptured aortic aneurysm in a defined and stable Swedish population was determined for the period 1952-1988. The annual rupture rate of abdominal aneurysm rose significantly from 0.9 per 100 000 inhabitants in the 1950s to 6.9 in the 1980s. After standardization for age, the mortality rate increased between 1960 and 1988 by 2.4 per cent annually (95 per cent confidence interval 1.2–3.6 per cent, P = 0.0004). Comparable figures from the UK during the 1980s have been reported to be two to three times higher. The mortality rate of ruptured thoracic aneurysm in the Swedish population did not increase when adjusted for age. In the 1980s the overall mortality rate of ruptured abdominal aneurysm was 85 per cent. Death occurred outside surgical clinics in 52 per cent of cases; 37 per cent of the total number of patients with ruptured aneurysm reached the operating table but only 30 per cent underwent aortic reconstruction. Patients treated by vascular surgeons had lower blood loss and transfusion needs, shorter aortic clamp time and operation time, and lower mortality rate than patients treated by non-specialized general surgeons. Specialized vascular surgeons also completed the reconstruction, and used straight grafts, in a higher proportion of cases than general surgeons.