Back to 2026 Abstracts
Individualized Growth In Abdominal Aortic Aneurysms: Insights From A Longitudinal Cohort
Jonathan Liao, BS, William Kartsonis, Ahmad Aljobeh, Stephanie Li, Praveen Parthasarathy, Apostolos Tassiopoulos.
Stony Brook University, Long Island, NY, USA.
OBJECTIVES: Abdominal aortic aneurysm (AAA) growth is commonly assumed to follow an exponential trajectory, yet direct evidence supporting this model is limited. Understanding whether growth is linear, exponential, or more heterogeneous is critical for risk prediction and surveillance planning. We investigated longitudinal growth patterns in a large patient cohort to test these assumptions and capture individual variability.
METHODS: We retrospectively analyzed 1,111 imaging studies from 419 patients (mean follow-up 6.3 ± 3.7 years). Each patient’s trajectory (AAA size vs. age) was fit with both linear and power regression (which allow for flexible rate changes over time) models. Model performance was compared using R² values. To evaluate acceleration, we compared initial and final growth rates derived from the nonlinear fits. At the population level, we plotted AAA size against corresponding growth rates for all 1,111 data points and analyzed mean growth across size quartiles (2.5-5.5 cm).
RESULTS: Power regression improved fit over linear regression in a subset of patients, but many were equally well described by linear models. Acceleration patterns varied widely: while some aneurysms demonstrated accelerating growth, others remained stable or even slowed over time, with a slight overall tendency toward positive acceleration (mean acceleration = .011 cm/yr2). At the population level, mean annual growth rates increased progressively across quartiles: 0.08, 0.14, 0.24, and 0.32 cm/year, respectively. Variability also increased with size, as reflected by rising standard deviations (0.09, 0.13, 0.17, and 0.22 cm/year).
CONCLUSIONS: AAA growth is more heterogeneous than the traditional exponential paradigm suggests. Although linear regression captured many individual trajectories, a subset demonstrated nonlinear acceleration or stabilizing decelerating behavior. Importantly, larger aneurysms exhibited not only faster average growth but also greater variability, highlighting their unpredictability. These findings underscore the need for personalized surveillance strategies and suggest that growth assumptions based on fixed models may oversimplify the clinical reality of AAA progression.
Back to 2026 Abstracts
