Ralf B Loeffler1, M Beth McCarville1, Anne W Wagstaff1,2,3, Matthew P Smeltzer4,5, Axel J Krafft1,6, Ruitian Song1, Jane S Hankins7, Claudia M Hillenbrand8. 1. Diagnostic Imaging, St. Jude Children's Research Hospital, 262 Danny Thomas Place, Memphis, TN, 38105-3678, USA. 2. Rhodes College, Memphis, TN, USA. 3. University of Alabama at Birmingham School of Medicine, Birmingham, AL, USA. 4. Department of Biostatistics, St. Jude Children's Research Hospital, Memphis, TN, USA. 5. Division of Epidemiology, Biostatistics, and Environmental Health, School of Public Health, University of Memphis, Memphis, TN, USA. 6. Department of Radiology, University Hospital Center Freiburg, Freiburg, Germany. 7. Department of Hematology, St. Jude Children's Research Hospital, Memphis, TN, USA. 8. Diagnostic Imaging, St. Jude Children's Research Hospital, 262 Danny Thomas Place, Memphis, TN, 38105-3678, USA. claudia.hillenbrand@stjude.org.
Abstract
BACKGROUND: Liver R2* values calculated from multi-gradient echo (mGRE) magnetic resonance images (MRI) are strongly correlated with hepatic iron concentration (HIC) as shown in several independently derived biopsy calibration studies. These calibrations were established for axial single-slice breath-hold imaging at the location of the portal vein. Scanning in multi-slice mode makes the exam more efficient, since whole-liver coverage can be achieved with two breath-holds and the optimal slice can be selected afterward. Navigator echoes remove the need for breath-holds and allow use in sedated patients. OBJECTIVE: To evaluate if the existing biopsy calibrations can be applied to multi-slice and navigator-controlled mGRE imaging in children with hepatic iron overload, by testing if there is a bias-free correlation between single-slice R2* and multi-slice or multi-slice navigator controlled R2*. MATERIALS AND METHODS: This study included MRI data from 71 patients with transfusional iron overload, who received an MRI exam to estimate HIC using gradient echo sequences. Patient scans contained 2 or 3 of the following imaging methods used for analysis: single-slice images (n = 71), multi-slice images (n = 69) and navigator-controlled images (n = 17). Small and large blood corrected region of interests were selected on axial images of the liver to obtain R2* values for all data sets. Bland-Altman and linear regression analysis were used to compare R2* values from single-slice images to those of multi-slice images and navigator-controlled images. RESULTS: Bland-Altman analysis showed that all imaging method comparisons were strongly associated with each other and had high correlation coefficients (0.98 ≤ r ≤ 1.00) with P-values ≤0.0001. Linear regression yielded slopes that were close to 1. CONCLUSION: We found that navigator-gated or breath-held multi-slice R2* MRI for HIC determination measures R2* values comparable to the biopsy-validated single-slice, single breath-hold scan. We conclude that these three R2* methods can be interchangeably used in existing R2*-HIC calibrations.
BACKGROUND: Liver R2* values calculated from multi-gradient echo (mGRE) magnetic resonance images (MRI) are strongly correlated with hepatic iron concentration (HIC) as shown in several independently derived biopsy calibration studies. These calibrations were established for axial single-slice breath-hold imaging at the location of the portal vein. Scanning in multi-slice mode makes the exam more efficient, since whole-liver coverage can be achieved with two breath-holds and the optimal slice can be selected afterward. Navigator echoes remove the need for breath-holds and allow use in sedated patients. OBJECTIVE: To evaluate if the existing biopsy calibrations can be applied to multi-slice and navigator-controlled mGRE imaging in children with hepatic iron overload, by testing if there is a bias-free correlation between single-slice R2* and multi-slice or multi-slice navigator controlled R2*. MATERIALS AND METHODS: This study included MRI data from 71 patients with transfusional iron overload, who received an MRI exam to estimate HIC using gradient echo sequences. Patient scans contained 2 or 3 of the following imaging methods used for analysis: single-slice images (n = 71), multi-slice images (n = 69) and navigator-controlled images (n = 17). Small and large blood corrected region of interests were selected on axial images of the liver to obtain R2* values for all data sets. Bland-Altman and linear regression analysis were used to compare R2* values from single-slice images to those of multi-slice images and navigator-controlled images. RESULTS: Bland-Altman analysis showed that all imaging method comparisons were strongly associated with each other and had high correlation coefficients (0.98 ≤ r ≤ 1.00) with P-values ≤0.0001. Linear regression yielded slopes that were close to 1. CONCLUSION: We found that navigator-gated or breath-held multi-slice R2* MRI for HIC determination measures R2* values comparable to the biopsy-validated single-slice, single breath-hold scan. We conclude that these three R2* methods can be interchangeably used in existing R2*-HIC calibrations.
Entities:
Keywords:
Children; Hepatic iron concentration; Iron; Iron quantification; Liver; Magnetic resonance imaging; R2* quantification; T2*
Authors: B Henninger; C Kremser; S Rauch; R Eder; H Zoller; A Finkenstedt; H J Michaely; M Schocke Journal: Eur Radiol Date: 2012-05-30 Impact factor: 5.315
Authors: M Beth McCarville; Claudia M Hillenbrand; Ralf B Loeffler; Matthew P Smeltzer; Ruitan Song; Chin-Shang Li; Jane S Hankins Journal: Pediatr Radiol Date: 2010-03-24
Authors: Utaroh Motosugi; Diego Hernando; Peter Bannas; James H Holmes; Kang Wang; Ann Shimakawa; Yuji Iwadate; Valentina Taviani; Jennifer L Rehm; Scott B Reeder Journal: J Magn Reson Imaging Date: 2015-04-01 Impact factor: 4.813
Authors: Latika Puri; Jonathan M Flanagan; Guolian Kang; Juan Ding; Wenjian Bi; Beth M McCarville; Ralf B Loeffler; Aaryani Tipirneni-Sajja; Martha Villavicencio; Kristine R Crews; Claudia M Hillenbrand; Jane S Hankins Journal: J Clin Med Date: 2019-11-05 Impact factor: 4.241