Rapid Fire Session

Session: Rapid Fire Session 1: Vascular, Coronary, and Cardiomyopathy (Monitor 9)

Reproducibility of centerline ascending aortic length measurements using image-based navigator contrast-enhanced magnetic resonance angiography

Thursday, February 5, 2026

10:50 AM - 10:57 AM

Location: Galapagos / Aruba / Ilhabela (2nd Lower Level)

Presenting Author(s)

JC

Jason Craft, MD

Cardiologist
St. Francis Hospital, The Heart Center
Greenvale, New York, United States

Primary Author(s)

JC

Jason Craft, MD

Cardiologist
St. Francis Hospital, The Heart Center
Greenvale, New York, United States

Co-Author(s)

AM

Amirhossein Moaddab, MD

Fellow of Advanced Cardiac Imaging
St. Francis Hospital, The Heart Center
Winston Salem, North Carolina, United States

Co-Author(s)

KP

Karli PIpitone, MPH

Research Analyst
St. Francis Hospital & Heart Center
Huntington, New York, United States
JW

Jonathan Weber, MPH

Research Analyst
St. Francis Hospital, The Heart Center
Greenvale, New York, United States
TC

Timothy Carter, MD

Cardiothoracic Surgeon
St. Francis Hospital & Heart Center, Good Samaritan Hospital
West Islip, New York, United States
MS

Michaela Schmidt

Applications Developer
Siemens Healthineers
Erlangen, Bayern, Germany
Karl P. Kunze, PhD

Senior Cardiac MR Scientist
Siemens Healthineers
Camberley, England, United Kingdom
RM

Ryan Murphy, BSc

Research Associate
St. Francis Hospital, The Heart Center, United States
Rene M. Botnar

Professor of Cardiovascular Imaging
Institute for Biological and Medical Engineering, Pontificia Universidad Católica de Chile
Santiago de Chile, Region Metropolitana, Chile
Claudia Prieto, PhD

Professor
Pontificia Universidad Católica de Chile / King's College London
Santiago, Region Metropolitana, Chile

Background: Ascending aortic length (AAL) is a predictor of type A aortic dissection (TAAD). Combining AAL of 110 mm and AA diameter of < 55 mm could preemptively identify 75% more of patients who develop TAAD compared to using the diameter cutoff alone. However, reproducible AAL measurement is contingent upon of proper identification of the aortic annulus, which is not possible with the ubiquitous first pass MRA method. Recently, image-based navigators (iNAV) for motion correction with100% respiratory efficiency and variable density Cartesian spiral-like (VD-CASPR) trajectory for acquisition acceleration, have enabled ECG gated contrast enhanced (CE)-MRA within a predictable scan time. We hypothesize that AAL can be measured reproducibly using ECG gated iNAV CE-MRA. We aim to determine the agreement of AAL within/between observers, and relationship between AAL, AA diameter and gender.

Methods: The previously described ECG gated iNAV CE-MRA research framework was utilized for inversion recovery gradient echo MRA at 1.5T (MAGNETOM Sola, Siemens Healthineers, Forchheim, Germany), and processed using TeraRecon Intuition, (TeraRecon, Durham, NC, USA). To generate the centerline, observers placed seeds at the following segments: sinus of Valsalva, mid AA, arch, proximal descending (DA), mid DA, hiatus, and abdominal aorta. The aortic annular plane was triangulated based on markers placed at the aortic cusp insertions. AAL was measured from the annulus to brachiocephalic artery (Fig 1). There were 31 subjects/40 MRA exams of patients with aortic disease. Intra-observer analysis was performed on physician 1 and & 2 AAL measurements for all MRAs and on aortic annular major/minor diameters, perimeter, and area performed by physician 2. AAL inter-observer comparison was also performed. Inter-class correlation coefficients (ICC) & 95% confidence intervals were estimated along with Bland-Altman plots and paired t-tests to examine reproducibility. We additionally performed linear regressions for AAL against maximum AA diameter (obtained by physician 2) and compared AAL within strata of gender for 25 patients without history of aortic repair.

Results: The mean age was 61±11 years and 23 (74%) were male. Near-perfect intra- and inter-observer correlations were observed for AAL (Fig 2A-C) as well as intra-observer measures of the aortic annular diameters, perimeter, and area by physician 2 with ICC values ranging from 0.95-0.98 (Table 1). A linear relationship was observed between maximum diameter and AAL (Figure 2D-E). Lastly, AAL was longer in males, significantly so when measured by physician 1, not by 2.

Conclusion: AAL, and aortic annulus measurements using ECG gated iNAV CE-MRA exhibit high reproducibility and replicability. The linear relationship between AAL and diameter is consistent with prior studies based on CTA-derived measurements, as well as AAL differences between genders.

Table 1. Intra and interobserver correlation coefficients estimating reproducibility of aortic dimensional measurements

Comparison	Measurement 1	Measurement 2	Mean Difference	p-value	ICC (95% CL)
Aortic Length (cm) Intraobserver Physician 1	10.51 ± 1.49	10.51 ± 1.42	0.01 ± 0.34	0.9112	0.97 (0.95, 0.99)
Aortic Length (cm) Intraobserver Physician 2	10.47 ± 1.48	10.41 ± 1.43	0.06 ± 0.28	0.2062	0.98 (0.96, 0.99)
Aortic Length (cm) Interobserver Physician 1 vs. 2	10.51 ± 1.49	10.47 ± 1.48	0.04 ± 0.38	0.5044	0.97 (0.94, 0.98)
Aortic Annular Major Diameter (cm) Intraobserver Physician 2	2.80 ± 0.29	2.80 ± 0.29	0.00 ± 0.09	0.9593	0.95 (0.91, 0.97)
Aortic Annular Minor Diameter (cm) Intraobserver Physician 2	2.23 ± 0.32	2.23 ± 0.31	0.00 ± 0.09	0.7889	0.96 (0.92, 0.98)
Aortic Annular Perimeter (cm) Intraobserver Physician 2	8.03 ± 0.95	7.98 ± 0.90	0.04 ± 0.26	0.2962	0.96 (0.93, 0.98)
Aortic Annular Area (cm2) Intraobserver Physician 2	4.95 ± 1.20	4.91 ± 1.16	0.04 ± 0.26	0.3567	0.97 (0.95, 0.99)
	Male	Female		p-value
Aortic Length (cm) by gender Physician 1	11.21 ± 1.02	9.59 ± 1.61		0.0388
Aortic Length (cm) by gender Physician 2	11.15 ± 1.08	9.70 ± 1.76		0.0776

Figure 1. Semi-automated centerline post-processing of contrast enhanced MRA images. The aortic annulus is defined by placing seeds at the insertion of the right, left and non-coronary cusp (A). Using the straightened multi-planar reformats, the ascending aortic length is measured from the aortic annulus to innominate artery origin (B).
Figure 2. Bland-Altman plots of the intraobserver analysis for aortic length measurements performed by physician 1 (A) and physician 2 (B), interobserver analysis (C) between aortic length measurements performed by physician 1 and physician 2 (N=31 subjects contributing 40 MRAs). Linear regression of maximum ascending aortic diameter and aortic length measured by physician 1 (D) and physician 2 (E) (N=25 subjects excluding aortic repair subjects and multiple MRAs).