TY - JOUR
T1 - Measurement of left atrial volume by 2D and 3D non-contrast computed tomography compared with cardiac magnetic resonance imaging
AU - Fredgart, Maise Høigaard
AU - Carter-Storch, Rasmus
AU - Møller, Jacob Eifer
AU - Øvrehus, Kristian Altern
AU - Pecini, Redi
AU - Dahl, Jordi Sanchez
AU - Gerke, Oke
AU - Alturkmany, Roudyna Ahmad
AU - Brandes, Axel
AU - Lindholt, Jes Sanddal
AU - Diederichsen, Axel Cosmus Pyndt
PY - 2018/4/19
Y1 - 2018/4/19
N2 - Background: Cardiac magnetic resonance imaging (MRI) is considered the gold standard for assessment of left atrial (LA) volume. We assessed the feasibility of evaluating LA volume using 3D non-contrast computed tomography (NCCT). Furthermore, since manual tracing of LA volume is time consuming, we evaluated the accuracy of the LA area using 2D NCCT imaging for LA volume assessment. Methods: MRI and NCCT imaging were performed in 69 patients before and one year after aortic valve replacement. In 3D MRI and 3D NCCT, each slice was manually traced, excluding the pulmonary veins and atrial appendage, and multiplied by slice spacing, thus generating a measure of LA volume. The LA volume was indexed to body surface area. On 2D NCCT, the largest axial cross-section LA area was traced manually. Results: The mean LA volume was 102 ± 28 ml in MRI compared with 103 ± 28 ml in 3D NCCT. 3D NCCT showed good agreement with MRI measurements (mean difference −0.7 ml/m2; 95% confidence interval (CI) −2.2 to 0.9). By Bland-Altman, 3D NCCT also showed good agreement with MRI (limits of agreement: −18.7–17.4 ml/m2). Furthermore, good correlation was found between 2D NCCT and 3D NCCT LA volume (r = 0.93). Conclusion: 2D and 3D measurements of LA volume in non-contrast computed tomography are feasible and accurate.
AB - Background: Cardiac magnetic resonance imaging (MRI) is considered the gold standard for assessment of left atrial (LA) volume. We assessed the feasibility of evaluating LA volume using 3D non-contrast computed tomography (NCCT). Furthermore, since manual tracing of LA volume is time consuming, we evaluated the accuracy of the LA area using 2D NCCT imaging for LA volume assessment. Methods: MRI and NCCT imaging were performed in 69 patients before and one year after aortic valve replacement. In 3D MRI and 3D NCCT, each slice was manually traced, excluding the pulmonary veins and atrial appendage, and multiplied by slice spacing, thus generating a measure of LA volume. The LA volume was indexed to body surface area. On 2D NCCT, the largest axial cross-section LA area was traced manually. Results: The mean LA volume was 102 ± 28 ml in MRI compared with 103 ± 28 ml in 3D NCCT. 3D NCCT showed good agreement with MRI measurements (mean difference −0.7 ml/m2; 95% confidence interval (CI) −2.2 to 0.9). By Bland-Altman, 3D NCCT also showed good agreement with MRI (limits of agreement: −18.7–17.4 ml/m2). Furthermore, good correlation was found between 2D NCCT and 3D NCCT LA volume (r = 0.93). Conclusion: 2D and 3D measurements of LA volume in non-contrast computed tomography are feasible and accurate.
KW - Aged
KW - Aortic Valve Stenosis/diagnostic imaging
KW - Aortic Valve/diagnostic imaging
KW - Body Surface Area
KW - Calcinosis/diagnostic imaging
KW - Feasibility Studies
KW - Female
KW - Heart Valve Prosthesis Implantation
KW - Humans
KW - Imaging, Three-Dimensional/methods
KW - Magnetic Resonance Imaging, Cine
KW - Male
KW - Middle Aged
KW - Multidetector Computed Tomography/methods
KW - Observer Variation
KW - Predictive Value of Tests
KW - Prospective Studies
KW - Radiographic Image Interpretation, Computer-Assisted/methods
KW - Reproducibility of Results
KW - Severity of Illness Index
KW - Time Factors
KW - Treatment Outcome
U2 - 10.1016/j.jcct.2018.04.001
DO - 10.1016/j.jcct.2018.04.001
M3 - Journal article
C2 - 29666031
AN - SCOPUS:85045451480
SN - 1934-5925
VL - 12
SP - 316
EP - 319
JO - Journal of Cardiovascular Computed Tomography
JF - Journal of Cardiovascular Computed Tomography
IS - 4
ER -