Refining the upper limit of normal for the ascending aorta: In search of optimal criteria -- a large database study...

The cutoff for dilated mid-ascending aorta (mAA) is controversial and has several definitions. The present study was carried out to determine the prevalence of mAA dilation based on published definitions and to identify the optimal cutoff.

Echocardiographic studies of patients >15 years of age performed at a large tertiary care center over 4 years, n = 49,330, were retrospectively evaluated. Leading-edge-to-leading-edge technique was used to measure the mAA in diastole. Several cutoff criteria were included. In addition, we defined normals in our database as those who, after 28 causes of dilated aorta were excluded, were normal both clinically and echocardiographically (n = 2334).

The mean age was 64.2 ± 17.1 years, and 31.5% were men. The prevalence of dilated mAA based on absolute criteria with sex stratification varied between 17% and 23% and based on relative criteria (to age, body surface area, and sex) varied between 6% and 11%. It further decreased to 7.6% on the addition of narrow age stratification (10 year intervals) performed on normals in our database. The multivariate adjusted R² (for variation in mAA diameter) was 0.25 for age, decreasing to 0.12 for weight and 0.07 for sex and height.

The lowest prevalence of 7.6% probably represents the optimal cutoff for dilated mAA because it includes age, which explains most of the variation in mAA, in narrow (10 year) intervals only performed in our normals, which represents the largest sample size to date.