Article Text
PDF
Abstract
Background Patients with hereditary hemorrhagic telangiectasia (HHT) have a high likelihood (10–20%) of developing one or more brain AVMs in addition to lung and liver AVMs. Current guidelines recommend screening MRI/MRA in childhood and adulthood to determine if HHT patients have brain AVMs, and, therefore guide counseling regarding observation or treatment. As part of a comprehensive ferumoxytol-enhanced 3T MRI/MRA protocol that allows evaluation of multiple organs in a single examination—thus improving patient comfort and convenience—we have begun to evaluate the sensitivity of this novel scan for brain AVMs.
Methods Under an IRB-approved protocol, we have imaged 11 HHT patients at 3T (Siemens Skyra), of whom 4 had known brain AVMs based on gold-standard digital subtraction angiograms (DSA). Following informed consent, subjects were administered 3 mg/kg ferumoxytol by intravenous infusion over 15 minutes, and were monitored during subsequent MRI/MRA (at least 30 minutes after end of IV infusion). VIBE sequence was acquired at 3 resolutions for all subjects: (1) 1.0 mm isotropic (TE 1.3 ms, TR 2.9 ms, acquisition time 1 minute), (2) 0.8 mm isotropic (TE 1.6 ms, TR 3.3 ms, acquisition time 2 minutes), and (3) 0.5 mm isotropic (TE 1.9 ms, TR 4.3 ms, acquisition time 4 minutes). Images were reformatted in multiple planes and with multiple maximal intensity projection slab thicknesses.
Results All known brain AVMs were detected by 3T ferumoxytol-enhanced VIBE MRA (FeMR). Compared to gold-standard DSA, FeMR accentuated the principal draining veins of AVMs as opposed to the halo of associated AVM nidus vessels (figure 1). AVMs were most easily detected at low spatial resolution (1.0 mm) imaging, as the halo of small nidus vessels were more clearly delineated compared to low signal brain parenchymal background. Edge-enhanced high-resolution (0.5 mm) images accentuated angioarchitecture within the AVM nidus, but were more challenging to interpret in terms of nidus size given overall pixelated image appearance. Higher resolution images with increased echo times may also have had signal dropout due to T2* effects.
Conclusion In the first few patients enrolled in the AVIATOR study, 3T screening FeMR appears to identify all AVMs seen on DSA. Further studies are needed to determine if high-risk features (e.g., feeding artery aneurysms, nidus aneurysms, venous outflow stenoses) of such AVMs can be reliably identified as compared to DSA.
DSA compared to 3T VIBE FeMR acquired at 0.8 mm isotropic resolution and reformatted in 3 planes as 10 mm thin maximal intensity projections
Disclosures S. Hetts: 1; C; Department of Defense, National Institutes of Health, Siemens, Stryker, Route 92. 2; C; Imperative. 4; C; Filtro, ThrombX. 5; C; UCSF. 6; C; Penumbra. A. Gill: None. D. Langston: None. T. Lomax Truong: None. Y. Park: None. D. Saloner: None.