| Objective:To investigate the correlation between the spectroscopic metabolic indices and the pathological grading of gliomas and evaluate 1H—MRS imaging for preoperative grading of suspected gliomas, and to detect the best spectroscopic index to gain optimized targets using the technique of automated proton magnetic resonance spectroscopy (1H—MRS) imaging guided frameless stereotactic biopsy; A meta-analysis was performed to evaluate the diagnostic performance of MRS in differentiating high-grade gliomas (HGGs) from low-grade gliomas (LGGs).Methods:This study contains 2 parts.In the first section, we retrospectively analyzed 142 patients with diagnosis in doubt who were underwent the automated 1H—MRS imaging guided frameless stereotactic biopsy between July 2011 and January 2015, and selected the patients with diagnosis of gliomas according to pathology. Finally,57 patients were enrolled. Among them,32 patients were male and 25 patients were female. Their age range from 13 years to 77 years, and the median age was 47 years. The correlation of the metabolic indices in the region of biopsy targets and glioma grade were analyzed. Receiver-operating characteristic curve (ROC) of different metabolic indices was made respectively, and comparisons of metabolic indices in different grade were performed.In the second section, PubMed and Embase databases were systematically searched for relevant studies of glioma grading assessed by MRS up to Mar 27,2015. Based on the data from eligible studies, pooled sensitivity, specificity, diagnostic odds ratio and areas under summary receiver operating characteristic curve (SROC) of different metabolite ratios were obtained.Results:In the first section,65 samples were obtained accurately corresponding to the selected targets in biopsy procedures. According to the classification of WHO criteria,39 (60%) samples were classified as low-grade gliomas including 38 with WHO grade ? and 1 with WHO ? grade whereas 26 samples were classified as high-grade gliomas including 18(28%) with WHO grade ? and 8(12%) with WHO grade ?. The Cho/Cr and Cho/NAA ratio have a positive correlation with the pathological grading, otherwise NAA and Cr value have a negative correlation with the grade. The correlation between Cho/NAA ratio and tumor grading is the most robust. NAA value, Cho/NAA and Cho/Cr ratios have a significantly difference in grade ?, grade ? and grade IV gliomas, while Cho, Cr value and NAA/Cr ratio have no significantly difference. HGGs were differentiated from LGGs with 93.8% diagnostic accuracy,92.0% sensitivity and 95.0% specificity by use of a Cho/NAA ratio threshold of 3.62. Cho/NAA ratio is superior to Cho/Cr and NAA/Cr ratios.In the second part, thirty articles comprising a total sample size of 1228 patients were included in our meta-analysis. Quantitative synthesis of studies showed that the pooled sensitivity/specificity of Cho/Cr, Cho/NAA and NAA/Cr ratio was 0.75/0.60, 0.80/0.76 and 0.71/0.70 respectively. The area under the curve (AUC) of the SROC was 0.83,0.87 and 0.78 respectively.Conclusions:The automated 1H—MRS imaging guided frameless stereotactic biopsy has advantage in biopsy surgery, 1H—MRS imaging allows preoperative grading of gliomas. For 1H—MRS imaging guided biopsy, the target of the Cho/NAA ratio hot spot is the first choice in homogenous glioma, and the target of the Cho/NAA ratio lower hot spot or the Cho/Cr ratio hot spot is the first choice in the gliomas with significant heterogeneity. MRS has moderate diagnostic performance in distinguishing HGGs from LGGs using tumoral metabolite ratios including Cho/Cr, Cho/NAA and NAA/Cr ratios. Although there was no significant difference in AUCs between Cho/Cr and Cho/NAA groups, Cho/NAA ratio showed higher sensitivity and specificity than Cho/Cr ratio and NAA/Cr ratio. |
PDF Full Text Request