https://doi.org/10.5812/iranjradiol.21289

Accurate Quantification of Metabolite Ratio in Osteosarcoma Employing Clinical Magnetic Resonance Spectroscopy at 3T; A Comparative Study With Normal Tissue

authors:

avatar Shaghayegh Karimi Alavijeh 1 , avatar Mohammadali Oghabian 2 , avatar Hamidreza Saligheh Rad 2 , *

Institute for Advanced Medical Technologies (IAMT)
Tehran University of Medical Sciences, Tehran, IR Iran
Iranian Journal of Radiology: Vol.11, issue 30th Iranian Congress of Radiology; e21289
published online: February 28, 2014
article type: Research Article

how to cite: Karimi Alavijeh S, Oghabian M, Saligheh Rad H. Accurate Quantification of Metabolite Ratio in Osteosarcoma Employing Clinical Magnetic Resonance Spectroscopy at 3T; A Comparative Study With Normal Tissue. I J Radiol. 2014;11(30th Iranian Congress of Radiology):e21289. https://doi.org/10.5812/iranjradiol.21289.

Abstract

Background:

Proton magnetic resonance spectroscopy imaging (1H-MRSI) has recently attracted attention in musculoskeletal applications in general, and in osteosarcoma in particular. Acquired spectra in the MRSI procedure are usually corrupted due to low signal-to-noise ratio (SNR), as well as baseline and residual water contaminations, leading to quantification errors and; therefore, misinterpretation of the ratios calculated by the clinical scanners software.

Objectives:

The aim of this study is accurate quantification of Choline-to-Creatine (Cho/Cr) ratio elevating in osteosarcoma.

Patients and Methods:

Cho/Cr were obtained in 10 patients and 10 healthy volunteers at 3T (Siemens Tim-Trio) by MRSI (PRESS, TR/TE: 2500/135 ms) using a phase-array coil with water-suppression. An extra single voxel spectroscopy (SVS) without water-suppression was acquired to provide phase information for further Eddy-current correction (ECC). Multi-stage preprocessing was applied as follows: ECC, SNR-enhancement and water-removal. Subtract-QUEST-MRSI as a time-domain technique was employed to accurately quantify the metabolite ratios and to estimate the baseline. An optimal database for subtract-QUEST was achieved based on multiple trials evaluated by acceptable peak fitting and Cramer-Rao-Bound (CRB) including Alanine, Cho, Cr, Lactate, Lipid and water. Lipids at frequencies of 0.94 and 1.33 ppm were combined to increase accuracy of the Lipid estimation.

Results:

The average values of Cho/Cr for 10 patients and 10 volunteers were 0.448 + 0.055 and 0.324 + 0.030, respectively, and the meaningful change of 35% (P < 0.0001) between the two groups. Reliability of the quantitative results was estimated by employing CRB which was about 93.6% among the 10 volunteers.

Conclusions:

In this study, we managed to observe meaningful changes in the Cho/Cr ratio in osteosarcoma lesions versus the control group employing 1H-MRSI. The best database for subtract-QUEST-MRSI was optimized in osteosarcoma. Reliability of the quantitative results was about 93.6% among the 10 volunteers.

Full Text

Full text is available in PDF