Preoperative Subtyping of WHO Grade 1 Meningiomas Using a Single-Shot Ultrafast MR T2 Mapping
Zongye Li,Hongyan Zhang,Xiao Wang,Yijie Yang,Yue Zhang,Yuchuan Zhuang,Zhiliang Wei,Qinqin Yang,Eryuan Gao,Yong Zhang,Shuhui Cai,Zhong Chen,Congbo Cai,Jianfeng Bao,Jingliang Cheng
DOI: https://doi.org/10.1002/jmri.29183
IF: 4.4
2024-01-01
Journal of Magnetic Resonance Imaging
Abstract:Background Meningioma subtype is crucial in treatment planning and prognosis delineation, for grade 1 meningiomas. T2 relaxometry could provide detailed microscopic information but is often limited by long scanning times.Purpose To investigate the potential of T2 maps derived from multiple overlapping-echo detachment imaging (MOLED) for predicting meningioma subtypes and Ki-67 index, and to compare the diagnostic efficiency of two different region-of-interest (ROI) placements (whole-tumor and contrast-enhanced, respectively).Study Type Prospective.Phantom/Subjects A phantom containing 11 tubes of MnCl2 at different concentrations, eight healthy volunteers, and 75 patients with grade 1 meningioma.Field Strength/Sequence3 T scanner. MOLED, T2-weighted spin-echo sequence, T2-dark-fluid sequence, and postcontrast T1-weighted gradient echo sequence.Assessment Two ROIs were delineated: the whole-tumor area (ROI1) and contrast-enhanced area (ROI2). Histogram parameters were extracted from T2 maps. Meningioma subtypes and Ki-67 index were reviewed by a neuropathologist according to the 2021 classification criteria.Statistical Tests Linear regression, Bland-Altman analysis, Pearson's correlation analysis, independent t test, Mann-Whitney U test, Kruskal-Wallis test with Bonferroni correction, and multivariate logistic regression analysis with the P-value significance level of 0.05.Results The MOLED T2 sequence demonstrated excellent accuracy for phantoms and volunteers (Mean(diff) = -1.29%, SDdiff = 1.25% and Mean(diff )= 0.36%, SDdiff = 2.70%, respectively), and good repeatability for volunteers (average coefficient of variance = 1.13%; intraclass correlation coefficient = 0.877). For both ROI1 and ROI2, T2 variance had the highest area under the curves (area under the ROC curve = 0.768 and 0.761, respectively) for meningioma subtyping. There was no significant difference between the two ROIs (P = 0.875). Significant correlations were observed between T2 parameters and Ki-67 index (r = 0.237-0.374).Data Conclusion MOLED T2 maps can effectively differentiate between meningothelial, fibrous, and transitional meningiomas. Moreover, T2 histogram parameters were significantly correlated with the Ki-67 index.