Truace
Subscribe

Circulating Metalloproteinase Inhibitors and Artificial Intelligence-Derived Radiomics Predict Persistent Endoleak After Endovascular Abdominal Aortic Aneurysm Repair: A Multimodal Biological-Imaging Model

Objective Persistent endoleak after endovascular abdominal aortic aneurysm repair (EVAR) reflects ongoing aneurysm sac biological activity and remains insufficiently predicted by anatomical surveillance alone. We aimed to evaluate whether integration of circulating matrix metalloproteinases (MMPs), tissue inhibitors of metalloproteinases (TIMPs), and artificial intelligence (AI)-derived radiomic features from computed tomographic angiography (CTA) improves prediction of persistent endoleak. Methods We conducted…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-1735· cite & verify
Circulating Metalloproteinase Inhibitors and Artificial Intelligence-Derived Radiomics Predict Persistent Endoleak After Endovascular Abdominal Aortic Aneurysm Repair: A Multimodal Biological-Imaging Model
Endovascular-aneurysm-repair-in-emergent-ruptured-abdominal-aortic-aneurysm-with-a-real-hostile-1471-2482-14-11-S5 by Wu N, Liu C, Fu Q, Zeng R, Chen Y, Yang G, Liu B. CC BY 2.0 · https://creativecommons.org/licenses/by/2.0

In brief

Researchers studied 62 consecutive patients who had elective infrarenal endovascular repair of an abdominal aortic aneurysm between 2020 and 2024. They measured six blood markers before surgery and at one and twelve months, and analyzed CT angiography scans with a semi-automated radiomics pipeline. Persistent endoleak at twelve months occurred in 18 patients.

The work suggests that follow-up after aneurysm repair could move beyond anatomy alone toward a combined biological and imaging approach, possibly allowing earlier risk sorting and tailored surveillance. The model is early-stage, built on a small retrospective group and only internally validated, so its usefulness in routine care is unproven until prospective testing.

Main points

  1. Persistent endoleak occurred in 18 of 62 patients, or 29 percent.
  2. Patients with persistent endoleak had higher MMP-9, lower TIMP-1 and TIMP-2, and higher MMP-9/TIMP-1 ratios at one month.
  3. The radiomics pipeline included reproducibility testing and penalized feature selection.
  4. The logistic regression model used event-per-variable control and bootstrap internal validation.

The gain

Combining AI-derived CT radiomic features with circulating metalloproteinase markers improved risk stratification for persistent endoleak after EVAR compared with morphological surveillance alone.

The rundown

The study was a retrospective observational analysis of patients treated at a single series level between January 2020 and December 2024. Plasma MMP-1, MMP-2, MMP-9, MMP-10, TIMP-1 and TIMP-2 were sampled preoperatively and at one and twelve months after repair.

CT angiography data were processed with a standardized semi-automated radiomics workflow, and the primary endpoint was persistent endoleak at twelve months. The authors link the event to systemic proteolytic imbalance and imaging-detectable thrombus heterogeneity.

The authors say the approach could support personalised surveillance intervals and identify patients who may need closer imaging follow-up or intervention. They state that the findings warrant prospective validation.

Sources

  1. Peer-reviewedJournal of Endovascular Therapy2026-10-08

The debate