1351102 - Feasibility of non-contrast 3D MR Angiography on a commercial 0.55T system; comparison with MRA on 1.5T in patients.

MR angiography (MRA) is commonly used to measure and track the dimensions of the thoracic aorta in patients with aortic dilatation. Non-contrast MRA is especially useful in patients who are pregnant, have poor renal function, contrast allergies, or inadequate IV access (1). Low-field, ultra-wide bore MRI can provide many benefits, including accommodation of severely obese and claustrophobic patients (2).

The purpose of this study is to evaluate a prototype ECG-triggered, navigator-gated, 3D bSSFP non-contrast MRA sequence on a commercial 0.55T low-field system, and to compare image quality and aortic dimension measurements against images obtained at 1.5T in a small cohort of patients.

Methods: Six patients (mean age: 54 years, 3 males) clinically evaluated for dilated thoracic aorta at 1.5T (Magnetom Sola, Siemens Healthcare, Erlangen, Germany) were also scanned at 0.55T (Free.Max, Siemens Healthcare, Erlangen, Germany) using ECG-triggered, navigator-gated, 3D bSSFP non-contrast MRA (3, 4). The average days between the two scans was 33 days (SD = 15.9). The gradient performance differs between these systems (Sola 45 mT/m, 200 mT/m/ms; Free.Max 26 mT/m, 45 mT/m/ms), requiring modifications in sequence parameters as shown in Table 1.  Cross-sectional aortic dimensions were measured (SuiteHeart, Neosoft, LLC, Pewaukee, WI) and compared at the locations shown in Figure 1. The agreement of the measurements was evaluated by intra-class correlation coefficient (ICC), where ICC >0.9 was considered to be excellent agreement. Image quality was blindly evaluated by three raters on a five-point scale based on level of artifacts and blurring.

Results:

Diagnostic image quality was obtained at 0.55T in all patients. The average image quality score at 0.55T (3.4) was less than that at 1.5T (4.1), p= 0.03. However, despite the measurable difference in image quality, the aortic measurements were in excellent agreement between field strengths, as shown in Table 2. 

Conclusion:

This study demonstrates that a prototype ECG-triggered, navigator-gated, 3D bSSFP non-contrast MRA sequence running on an ultra-wide bore 0.55T low-field system with reduced gradient performance, provides vessel dimension measurements in agreement with images acquired at 1.5T. While respiratory artifacts contributed to reduced image quality scores at 0.55T, the vessel dimension measurements were still in agreement within each patient and at each anatomical location. The present study demonstrates feasibility of non-contrast MRA at 0.55T; future work will evaluate diagnostic performance of the technique in patients with severe obesity and claustrophobia.