Feasibility and Validity of a Standardized Protocol for Measuring Thoracic Aortic Aneurysm Morphology

Thoracic endovascular aortic repair (TEVAR) is the predominant treatment for thoracic aortic aneurysms (TAA) due to its superior perioperative outcomes. International guidelines recommend assessing TAA characteristics via computed tomography angiography (CTA). However, the lack of a standardized measurements protocol introduces variability in preoperative planning, and imaging surveillance, and therefore hinders artificial intelligence (AI) integration for fully automated measurements. This study aims to develop and validate a standardized measurement protocol for TAA to improve consistency in measurement and enhance imaging surveillance.

A retrospective cohort study was performed at a Dutch tertiary center on patients who underwent TEVAR for a descending TAA from 2010 to 2019. We included degenerative and mycotic TAAs, and exclusions were the lack of a preoperative CTA and/or incomplete postoperative imaging. A standardized measurement protocol was developed based on expert opinion and validated endovascular aortic repair (EVAR) protocols. Imaging analysis utilized dedicated 3D imaging software. The protocol included semi-automated 3D segmentation, center lumen line (CLL) reconstruction, and several measurements, including aortic diameter/volume and sealing lengths. Intraobserver and interobserver agreements were assessed using Bland-Altman analysis and intraclass correlation coefficients (ICCs).

We analyzed 133 CTA scans from 31 patients, showing high levels of agreement across measurements, particularly for those repeated by the same observer. Maximum diameter measurements demonstrated excellent consistency, with minimal mean differences for intraobserver and interobserver agreements and excellent correlation (ICC>0.900). Volume measurements were similar consistent, with mean differences of 3.45 (intraobserver) and 2.75 (interobserver) cc. Proximal and distal seal measurements showed good agreement, although interobserver correlation was slightly less consistent. Length of coverage by the endograft exhibited strong consistency.

Our standardized measurement protocol for descending TAA offers a consistent approach for preoperative planning, imaging surveillance, and research applications, with high agreement in most measurements. This consistency could reduce variability, enhance imaging surveillance. Future research should focus on external validation, integrating AI to further improve measurement consistency, and simplify TEVAR surveillance.Clinical ImpactThis retrospective study of 133 CTA scans from 31 thoracic endovascular aortic repair (TEVAR) for TAA patients demonstrated excellent measurements consistency, including maximum diameter showing strong agreement and minimal differences in volume measurements. In the absence of an existing measurements protocol, our proposed standardized protocol supports improved preoperative planning, imaging surveillance to further enhance TEVAR surveillance.