OBJECTIVES: The aim of this study was to compare the reproducibility of 7 late gadolinium enhancement (LGE) quantification techniques across 3 conditions in which LGE is known to be important: acute myocardial infarction (AMI), chronic myocardial infarction (CMI), and hypertrophic cardiomyopathy (HCM). BACKGROUND: LGE by cardiac magnetic resonance is the gold-standard technique for assessing myocardial scar. No consensus exists on the best method for its quantification, and research in this area is scant. Techniques include manual quantification, thresholding by 2, 3, 4, 5, or 6 SDs above remote myocardium, and the full width at half maximum (FWHM) technique. To date, LGE has been linked to outcome in 3 conditions: AMI, CMI, and HCM. METHODS: Sixty patients with 3 LGE etiologies (AMI, n = 20; CMI, n = 20; HCM, n = 20) were scanned for LGE. LGE volume was quantified using the 7 techniques. Mean LGE volume, interobserver and intraobserver reproducibility, and impact on sample size were assessed. RESULTS: LGE volume varied significantly with the quantification method used. There was no statistically significant difference between LGE volume by the FWHM, manual, and 6-SD or 5-SD techniques. The 2-SD technique generated LGE volumes up to 2 times higher than the FWHM, 6-SD, and manual techniques. The reproducibility of all techniques was worse in HCM than AMI or CMI. The FWHM technique was the most reproducible in all 3 conditions compared with any other method (p < 0.001). Use of the FWHM technique for LGE quantification in paired analysis would lead to at least a 60% reduction in required sample size compared with any other method. CONCLUSIONS: Regardless of the disease under study, the FWHM technique for LGE quantification gives LGE volume mean results similar to manual quantification and is statistically the most reproducible, reducing required sample sizes by up to one-half.
OBJECTIVES: The aim of this study was to compare the reproducibility of 7 late gadolinium enhancement (LGE) quantification techniques across 3 conditions in which LGE is known to be important: acute myocardial infarction (AMI), chronic myocardial infarction (CMI), and hypertrophic cardiomyopathy (HCM). BACKGROUND: LGE by cardiac magnetic resonance is the gold-standard technique for assessing myocardial scar. No consensus exists on the best method for its quantification, and research in this area is scant. Techniques include manual quantification, thresholding by 2, 3, 4, 5, or 6 SDs above remote myocardium, and the full width at half maximum (FWHM) technique. To date, LGE has been linked to outcome in 3 conditions: AMI, CMI, and HCM. METHODS: Sixty patients with 3 LGE etiologies (AMI, n = 20; CMI, n = 20; HCM, n = 20) were scanned for LGE. LGE volume was quantified using the 7 techniques. Mean LGE volume, interobserver and intraobserver reproducibility, and impact on sample size were assessed. RESULTS: LGE volume varied significantly with the quantification method used. There was no statistically significant difference between LGE volume by the FWHM, manual, and 6-SD or 5-SD techniques. The 2-SD technique generated LGE volumes up to 2 times higher than the FWHM, 6-SD, and manual techniques. The reproducibility of all techniques was worse in HCM than AMI or CMI. The FWHM technique was the most reproducible in all 3 conditions compared with any other method (p < 0.001). Use of the FWHM technique for LGE quantification in paired analysis would lead to at least a 60% reduction in required sample size compared with any other method. CONCLUSIONS: Regardless of the disease under study, the FWHM technique for LGE quantification gives LGE volume mean results similar to manual quantification and is statistically the most reproducible, reducing required sample sizes by up to one-half.
Authors: Tomas G Neilan; Ravi V Shah; Siddique A Abbasi; Hoshang Farhad; John D Groarke; John A Dodson; Otavio Coelho-Filho; Ciaran J McMullan; Bobak Heydari; Gregory F Michaud; Roy M John; Rob van der Geest; Michael L Steigner; Ron Blankstein; Michael Jerosch-Herold; Raymond Y Kwong Journal: J Am Coll Cardiol Date: 2013-08-28 Impact factor: 24.094
Authors: Pascale Beliveau; Farida Cheriet; Stasia A Anderson; Joni L Taylor; Andrew E Arai; Li-Yueh Hsu Journal: Comput Biol Med Date: 2015-08-08 Impact factor: 4.589
Authors: E Gerbaud; H Cochet; E Bullier; C Ragot; S H Gilbert; H Douard; Y Pucheu; F Laurent; P Coste; L Bordenave; M Montaudon Journal: Br J Radiol Date: 2014-04-29 Impact factor: 3.039