Synopsis
Applying vessel size imaging(VSI) and dynamic susceptibility contrast MR
(DSC-MR) to explore difference of
the microvessel between grade Ⅱ and Ⅲ oligodendroglioma. We discovered both VSI and
DSC-MR could identify different grades of oligodendroglioma, but the power of
VSI was better than DSC-MR.
Purpose
To explore
difference of the microvessel between grade Ⅱ and Ⅲ oligodendroglioma via vessel size
imaging(VSI) and DSC-MR.
Method
We
retrospectively analyzed 36 cases oligodendroglioma patients（22 gradeⅡ and 14 grade Ⅲ tumors）confirmed
by surgery. 18 patients were undertook
multi-GE-SE sequence to
generate VSI
maps by GE Signa HDX 1.5T MR scanner, and 12 cases were accepted DSC-MR to obtain cerebral blood volume(CBV) maps by Siemens 3.0T MR scanner, 6 patients were underwent both methods. To evaluate the VSI and CBV maps by hot spot method in tumor region obtaining mean VSI(VSImean), maximum VSI(VSImax), mean relative CBV(rCBVmean), maximum relative CBV(rCBVmax). CD34 was a mark of microvascular caliber (VShis) in VSI scanning patients of paraffin sections．Pearson correlation
analysis was used to evaluate the corelation
between VShis
and VSImean, VSImax
respectively．Independent Student’s t-test was used to inter-group comparisons．ROC analysis was
used to evaluate the value of these two
imaging technologies in distinguishing gradeⅡ and Ⅲ oligodendroglioma．Results
A total of 24 patients with VSI(13 gradeⅡ and 11 grade Ⅲ tumors), and 18 patients with DSC-MR(12 gradeⅡ and 6 grade Ⅲ tumors). Both VSImean and VSImax were strongly correlation with VShis, correlation coefficient were
0.747
and 0.701
respectively. In gradeⅡ and Ⅲ oligodendrogliomas, VSImean were
(38.8±18.3)μm
and (89.9±35.5μm, VSImax were (47.4±16.5)μm
and (112.4±38.9)μm,
VShis were
(8.5±2.0)μm and (12.7±4.8)μm, rCBVmax were (4.8±2.1) and (6.6±1.6), rCBVmean were
(3.9±2.0) and (5.9±1.4）respectively．VSImean, VSImax, VShis and rCBVmean showed significant difference (t=-4.54、-5.50、-2.87、-2.18，p
<0.05) between gradeⅡ and Ⅲ
oligodendroglioma, while no significant
difference showed in rCBVmax（t=-1.87，p=0.079）．ROC analysis revealed
that the optimal cutoff
value, sensitivity, specificity and AUC of
VSImean were 52.6μm, 100%,
92%,
0.97
respectively; VSImax were 81.2μm, 100%, 100%, respectively; rCBVmean were 5.4, 83%, 83%, 0.79 respectively, while rCBVmax were
6.2, 83%,
83%,
0.78
respectively. Conclusion
Compared with
histology, VSI can accurately access microvascular caliber of oligodendroglioma. Furthermore, VSI also can identify different
grades better than DSC-MR for its sensitivity, specific and accuracy. In
conclusion, VSI can provide information for preoperatively
assessment oligodendroglioma grade.
Acknowledgements
No acknowledgement found.References
1. Tropres I, Grimault S, Vaeth A, et al. Vessel size
imaging[J]. Magn Reson Med, 2001, 45(3):397-408.
2.
Kiselev VG, Strecker R,
Ziyeh S, et al. Vessel size imaging in humans[J]. Magn Reson Med, 2005, 53(3):553-563.
3.
Kang HY, Xiao HL, Chen
JH, et al. Comparison of the Effect of Vessel Size Imaging and Cerebral Blood
Volume Derived from Perfusion MR Imaging on Glioma Grading[J]. AJNR Am J
Neuroradiol, 2016, 37(1):51-57.
4. Lemasson B, Valable S, Farion R, et al. In vivo imaging of
vessel diameter, size, and density: a comparative study between MRI and
histology[J]. Magn Reson Med, 2013, 69(1):18-26
5.
5. Kellner E, Breyer T, Gall P, et al. MR evaluation of vessel
size imaging of human gliomas: Validation by histopathology[J]. J Magn Reson
Imaging, 2015, 42(4):1117-1125.