Cargando…

Gadolinium free cardiovascular magnetic resonance with 2-point Cine balanced steady state free precession

BACKGROUND: Cardiovascular magnetic resonance (CMR) of ventricular structure and function is widely performed using cine balanced steady state free precession (bSSFP) MRI. The bSSFP signal of myocardium is weighted by magnetization transfer (MT) and T1/T2-relaxation times. In edematous and fibrotic...

Descripción completa

Detalles Bibliográficos
Autores principales: Stromp, Tori A., Leung, Steve W., Andres, Kristin N., Jing, Linyuan, Fornwalt, Brandon K., Charnigo, Richard J., Sorrell, Vincent L., Vandsburger, Moriel H.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: BioMed Central 2015
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4628395/
https://www.ncbi.nlm.nih.gov/pubmed/26520782
http://dx.doi.org/10.1186/s12968-015-0194-1
Descripción
Sumario:BACKGROUND: Cardiovascular magnetic resonance (CMR) of ventricular structure and function is widely performed using cine balanced steady state free precession (bSSFP) MRI. The bSSFP signal of myocardium is weighted by magnetization transfer (MT) and T1/T2-relaxation times. In edematous and fibrotic tissues, increased T2 and reduced MT lead to increased signal intensity on images acquired with high excitation flip angles. We hypothesized that acquisition of two differentially MT-weighted bSSFP images (termed 2-point bSSFP) can identify tissue that would enhance with gadolinium similar to standard of care late gadolinium enhancement (LGE). METHODS: Cine bSSFP images (flip angles of 5° and 45°) and native-T1 and T2 maps were acquired in one mid-ventricular slice in 47 patients referred for CMR and 10 healthy controls. Afterwards, LGE images and post-contrast T1 maps were acquired and gadolinium partition coefficient (GPC) was calculated. Maps of ΔS/S(o) were calculated as (S(45)-S(5))/S(5)*100 (%), where S(flip_angle) is the voxel signal intensity. RESULTS: Twenty three patients demonstrated areas of myocardial hyper-enhancement with LGE. In enhanced regions, ΔS/S(o), native-T1, T2, and GPC were heightened (p < 0.05 vs. non-enhanced tissues). ΔS/S(o), native-T1, and T2 all demonstrated association with GPC, however the association was strongest for ΔS/S(o). Bland-Altman analysis revealed a slight bias towards larger volume of enhancement with ΔS/S(o) compared to LGE, and similar transmurality. Subjective analysis with 2-blinded expert readers revealed agreement between ΔS/S(o) and LGE of 73.4 %, with false positive detection of 16.7 % and false negative detection of 15.2 %. CONCLUSIONS: Gadolinium free 2-point bSSFP identified tissue that enhances at LGE with strong association to GPC. Our results suggest that with further development, MT-weighted CMR could be used similar to LGE for diagnostic imaging. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12968-015-0194-1) contains supplementary material, which is available to authorized users.